
#data science packages
import pandas as pd
import numpy as np

#visualisation packages
import plotly.io as pio
import plotly.graph_objects as go
import pyreadstat
import matplotlib.pyplot as plt
import matplotlib.colorbar as mcb
import seaborn as sns

#packages needed to plot on a map
from adjustText import adjust_text
import pyogrio
import geopandas as gpd
import pycountry


pio.renderers.default='notebook'
import plotly.offline as pyo
pyo.init_notebook_mode()


df, meta = pyreadstat.read_sav("lrf_wrp_2021_full_data.sav")
df.head()


df_nepal = df[df.Country=='Nepal']
df_nepal.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 2095 entries, 107875 to 280087
Columns: 208 entries, WPID_RANDOM to Q5F_2019
dtypes: float64(207), object(1)
memory usage: 3.3+ MB


#filter out countries with number of data points less than 1200
df = df.groupby('Country').filter(lambda x: x['Country'].count() >= 1200)

df_1=df[df['Q1'].isnull()]


chart = df['Country'].value_counts().plot(kind='bar',
                                    figsize=(30,8),
                                    title="Number of participants from each Country")
chart.set_xlabel("Country")
chart.set_ylabel("Number of respondents")
plt.xticks(rotation = 90)
plt.show()


plt.figure(figsize=(10,4))
chart=sns.countplot(x='GlobalRegion', data=df);
chart.set_xticklabels(chart.get_xticklabels(), rotation=45, horizontalalignment='right')
chart.set_xticks([0,1,2,3,4,5,6,7,8,9,10,11,12,13,14])
chart.set_xticklabels(['Eastern Africa','Central/Western Africa','North Africa','Southern Africa','Latin America & Caribbean','Northern America','Central Asia','East Asia','South-eastern Asia','South Asia','Middle East','Eastern Europe','Northern/Western Europe','Southern Europe','Australia and New Zealand'])
chart.set_title('Number of participants from each Region')
plt.show()


# Define the mapping of values to labels
age_labels = {1: '15-29', 2: '30-49', 3: '50-64', 4: '65+'}
gender_labels = {1: 'Male', 2: 'Female'}

# Create the plot
fig, axes = plt.subplots(1,3,figsize=(20,4))
#chart = sns.countplot(x='AgeGroups4', data=df, hue='Gender', hue_order=[1, 2], width=0.4, ax=axes[0])
chart = sns.countplot(x='AgeGroups4', data=df, hue='Gender', hue_order=[1, 2], ax=axes[0])
sns.move_legend(chart, "upper right", title='Gender')
chart.set_xticks([0, 1, 2, 3])
chart.set_title("Gender vs Age category")

# Rename the x-axis tick labels
new_labels = [age_labels[val] for val in range(1, 5)]
chart.set_xticklabels(new_labels)

# Rename the legend labels
handles, labels = axes[0].get_legend_handles_labels()
new_labels = [gender_labels[int(label)] for label in labels]
axes[0].legend(handles, new_labels, title='Gender')


grouped = df.groupby(['AgeGroups4', 'Gender']).size().reset_index(name='count')
totals = grouped.groupby('AgeGroups4')['count'].sum().reset_index(name='total')
grouped = pd.merge(grouped, totals, on='AgeGroups4')
grouped['percentage'] = grouped['count'] / grouped['total'] * 100

# Create the plot
#chart = sns.barplot(x='AgeGroups4', y='percentage', hue='Gender', data=grouped, hue_order=[1, 2], width=0.4,ax=axes[1])
chart = sns.barplot(x='AgeGroups4', y='percentage', hue='Gender', data=grouped, hue_order=[1, 2], ax=axes[1])

sns.move_legend(chart, "upper right", title='Gender')
chart.set_xticklabels([age_labels[val] for val in range(1, 5)])
chart.set_title("Percentage by Age category")

# Rename the legend labels
handles, labels = axes[1].get_legend_handles_labels()
new_labels = [gender_labels[int(label)] for label in labels]
axes[1].legend(handles, new_labels, title='Gender')

grouped_1 = df.groupby(['Gender', 'AgeGroups4']).size().reset_index(name='count')
totals = grouped_1.groupby('Gender')['count'].sum().reset_index(name='total')
grouped_1 = pd.merge(grouped_1, totals, on='Gender')
grouped_1['percentage'] = grouped_1['count'] / grouped_1['total'] * 100

# Create the plot
#chart = sns.barplot(x='Gender', y='percentage', hue='AgeGroups4', data=grouped_1, hue_order=[1, 2,3,4], width=0.4,ax=axes[2])
chart = sns.barplot(x='Gender', y='percentage', hue='AgeGroups4', data=grouped_1, hue_order=[1, 2,3,4], ax=axes[2])

sns.move_legend(chart, "upper right", title='AgeGroups4')

chart.set_xticklabels([gender_labels[val] for val in range(1, 3)])
chart.set_title("Percentage by Gender")

# Rename the legend labels
handles, labels = axes[2].get_legend_handles_labels()
new_labels = [age_labels[int(label)] for label in labels]
axes[2].legend(handles, new_labels, title='Age_Groups')

# Show the plot
plt.show()


# Define the mapping of values to labels
edu_labels = {1: 'Primary', 2: 'Secondary', 3: 'Tertiary', 4: 'DK/Refused'}
income_labels = {1: 'Poorest 20%', 2: 'Second 20%', 3: 'Middle 20%', 4: 'Fourth 20%', 5: 'Richest 20%'}

# Create the plot
fig, axes = plt.subplots(1,3,figsize=(20,4))
#chart = sns.countplot(x='Education', data=df, hue='INCOME_5', hue_order=[1, 2,3,4,5], width=0.4, ax=axes[0])
chart = sns.countplot(x='Education', data=df, hue='INCOME_5', hue_order=[1, 2,3,4,5], ax=axes[0])

sns.move_legend(chart, "upper right", title='Per Capita Income Quintiles')
chart.set_xticks([0, 1, 2, 3])
chart.set_title("Education vs Per Capita Income Quintiles")

# Rename the x-axis tick labels
new_labels = [edu_labels[val] for val in range(1, 5)]
chart.set_xticklabels(new_labels)

# Rename the legend labels
handles, labels = axes[0].get_legend_handles_labels()
new_labels = [income_labels[int(label)] for label in labels]
axes[0].legend(handles, new_labels, title='Per Capita Income Quintiles')


grouped = df.groupby(['Education', 'INCOME_5']).size().reset_index(name='count')
totals = grouped.groupby('Education')['count'].sum().reset_index(name='total')
grouped = pd.merge(grouped, totals, on='Education')
grouped['percentage'] = grouped['count'] / grouped['total'] * 100

# Create the plot
chart = sns.barplot(x='Education', y='percentage', hue='INCOME_5', data=grouped, hue_order=[1, 2,3,4,5], ax=axes[1])
sns.move_legend(chart, "upper right", title='Per Capita Income Quintiles')
chart.set_xticklabels([edu_labels[val] for val in range(1, 5)])
chart.set_title("Percentage by Education")

# Rename the legend labels
handles, labels = axes[1].get_legend_handles_labels()
new_labels = [income_labels[int(label)] for label in labels]
axes[1].legend(handles, new_labels, title='Per Capita Income Quintiles')

grouped_1 = df.groupby(['INCOME_5', 'Education']).size().reset_index(name='count')
totals = grouped_1.groupby('INCOME_5')['count'].sum().reset_index(name='total')
grouped_1 = pd.merge(grouped_1, totals, on='INCOME_5')
grouped_1['percentage'] = grouped_1['count'] / grouped_1['total'] * 100

# Create the plot
chart = sns.barplot(x='INCOME_5', y='percentage', hue='Education', data=grouped_1, hue_order=[1, 2,3,4], ax=axes[2])
sns.move_legend(chart, "upper right", title='Education')
chart.set_xticklabels([income_labels[val] for val in range(1, 6)])
chart.set_title("Percentage by Per Capita Income Quintiles")

# Rename the legend labels
handles, labels = axes[2].get_legend_handles_labels()
new_labels = [edu_labels[int(label)] for label in labels]
axes[2].legend(handles, new_labels, title='Education')

# Show the plot
plt.show()


# Define the mapping of values to labels
safety_labels = {1: 'More safe', 2: 'Less safe', 3: 'About as safe', 98: 'DK', 99: 'Refused'}
age_labels = {1: '15-29', 2: '30-49', 3: '50-64', 4: '65+'}
gender_labels = {1: 'Male', 2: 'Female'}
housesize_labels = {1: '1-2 people', 2: '3-4 people', 3: '5-9 people', 4: '10 or more'}
urban_labels = {1: 'Rural', 2: 'City', 9: 'DK/Refused'}
income_labels = {1: 'Poorest 20%', 2: 'Second 20%', 3: 'Middle 20%', 4: 'Fourth 20%', 5: 'Richest 20%'}

# Create the plot
fig, axes = plt.subplots(2,3,figsize=(20,6))

chart=sns.countplot(x='Q1', data=df,  ax=axes[0,0]);
chart.set_xticks([0,1,2,3,4])
chart.set_title("safety compared with five years ago")
# Set the x-axis tick labels
new_labels = [safety_labels[val] if val in safety_labels else "N/A" for val in range(1, 6)]
chart.set_xticklabels(new_labels)
chart.set_xlabel('Safety')


grouped = df.groupby(['AgeGroups4', 'Q1']).size().reset_index(name='count')
totals = grouped.groupby('AgeGroups4')['count'].sum().reset_index(name='total')
grouped = pd.merge(grouped, totals, on='AgeGroups4')
grouped['percentage'] = grouped['count'] / grouped['total'] * 100
# Create the plot
chart = sns.barplot(x='AgeGroups4', y='percentage', hue='Q1', data=grouped, hue_order=[1, 2,3,98,99], ax=axes[0,1])
sns.move_legend(chart, "upper right", title='Safety')
chart.set_xticklabels([age_labels[val] for val in range(1, 5)])
chart.set_title("Percentage by Age Group")
chart.set_xlabel('Age Group')
# Rename the legend labels
handles, labels = axes[0,1].get_legend_handles_labels()
new_labels = [safety_labels[int(label)] for label in labels]
axes[0,1].legend(handles, new_labels, title='Safety')

grouped = df.groupby(['Gender', 'Q1']).size().reset_index(name='count')
totals = grouped.groupby('Gender')['count'].sum().reset_index(name='total')
grouped = pd.merge(grouped, totals, on='Gender')
grouped['percentage'] = grouped['count'] / grouped['total'] * 100
# Create the plot
chart = sns.barplot(x='Gender', y='percentage', hue='Q1', data=grouped, hue_order=[1, 2,3,98,99], ax=axes[0,2])
sns.move_legend(chart, "upper right", title='Safety')
chart.set_xticklabels([gender_labels[val] for val in range(1, 3)])
chart.set_title("Percentage by Gender")
chart.set_xlabel('Gender')
# Rename the legend labels
handles, labels = axes[0,2].get_legend_handles_labels()
new_labels = [safety_labels[int(label)] for label in labels]
axes[0,2].legend(handles, new_labels, title='Safety')

grouped = df.groupby(['HouseholdSize', 'Q1']).size().reset_index(name='count')
totals = grouped.groupby('HouseholdSize')['count'].sum().reset_index(name='total')
grouped = pd.merge(grouped, totals, on='HouseholdSize')
grouped['percentage'] = grouped['count'] / grouped['total'] * 100
# Create the plot
chart = sns.barplot(x='HouseholdSize', y='percentage', hue='Q1', data=grouped, hue_order=[1, 2,3,98,99],ax=axes[1,0])
sns.move_legend(chart, "upper right", title='Safety')
chart.set_xticklabels([housesize_labels[val] for val in range(1, 5)])
chart.set_title("Percentage by Household Size")
chart.set_xlabel('Household Size')
# Rename the legend labels
handles, labels = axes[1,0].get_legend_handles_labels()
new_labels = [safety_labels[int(label)] for label in labels]
axes[1,0].legend(handles, new_labels, title='Safety')

grouped = df.groupby(['Urbanicity', 'Q1']).size().reset_index(name='count')
totals = grouped.groupby('Urbanicity')['count'].sum().reset_index(name='total')
grouped = pd.merge(grouped, totals, on='Urbanicity')
grouped['percentage'] = grouped['count'] / grouped['total'] * 100
# Create the plot
chart = sns.barplot(x='Urbanicity', y='percentage', hue='Q1', data=grouped, hue_order=[1, 2,3,98,99], ax=axes[1,1])
sns.move_legend(chart, "upper right", title='Safety')
chart.set_xticklabels([urban_labels[val] if val in urban_labels else "N/A" for val in range(1, 4)])
chart.set_title("Percentage by Household Size")
chart.set_xlabel('Household Size')
# Rename the legend labels
handles, labels = axes[1,1].get_legend_handles_labels()
new_labels = [safety_labels[int(label)] for label in labels]
axes[1,1].legend(handles, new_labels, title='Safety')

grouped = df.groupby(['INCOME_5', 'Q1']).size().reset_index(name='count')
totals = grouped.groupby('INCOME_5')['count'].sum().reset_index(name='total')
grouped = pd.merge(grouped, totals, on='INCOME_5')
grouped['percentage'] = grouped['count'] / grouped['total'] * 100
# Create the plot
chart = sns.barplot(x='INCOME_5', y='percentage', hue='Q1', data=grouped, hue_order=[1, 2,3,98,99],ax=axes[1,2])
sns.move_legend(chart, "upper right", title='Safety')
chart.set_xticklabels([income_labels[val] for val in range(1, 6)])
chart.set_title("Percentage by Household Size")
chart.set_xlabel('Household Size')
# Rename the legend labels
handles, labels = axes[1,2].get_legend_handles_labels()
new_labels = [safety_labels[int(label)] for label in labels]
axes[1,2].legend(handles, new_labels, title='Safety')

# Adjust horizontal space between subplots
fig.subplots_adjust(hspace=0.5)

# Show the plot
plt.show()


sns.set(font_scale=1)
#fig, axes = plt.subplots(figsize=(12,4))

#fig.suptitle('Feel More, Less or About as Safe Compared With Five Years Ago')
g1=sns.countplot(x='GlobalRegion', data=df, hue='Q1');
g1.legend(loc='upper center',ncol=2, labels=['Safer','Less safe','About as safe','Do not know','Refused'])
#sns.move_legend(g1, "upper right", title='Safety', labels=['More safe','Less safe','About as safe','DK','Refused'])
g1.set_xticklabels(g1.get_xticklabels(), rotation = 90)
g1.set_xticks([0,1,2,3,4,5,6,7,8,9,10,11,12,13,14])
g1.set_xticklabels(['Eastern Africa','Central/Western Africa','North Africa','Southern Africa','Latin America & Caribbean','Northern America','Central Asia','East Asia','South-eastern Asia','South Asia','Middle East','Eastern Europe','Northern/Western Europe','Southern Europe','Australia and New Zealand'])
g1.set_xlabel('')
g1.set_ylabel('Count')


plt.savefig('Figures/safety_region.pdf',bbox_inches="tight")
plt.show()


plt.figure(figsize=(10,4))
chart=sns.countplot(x='Q2_1', data=df);
chart.set_xticklabels(chart.get_xticklabels(), rotation=90, horizontalalignment='right')
chart.set_xlabel('Risks')
chart.set_xticks([0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24])
chart.set_xticklabels(['TRANS1','TRANS2','CRIME','WAR','HEALTH1','HEALTH2','HEALTH3','HEALTH4','FINANCIAL','ECONOMY','POLITICS','TECHNOLOGY','WATER','FOOD1','FOOD2','COOKING','WORK','ENVIRONMENT1','ENVIRONMENT2','ENVIRONMENT3','DROWNING','Other','Nothing','DK','Refused'])
chart.set_title('Greatest Source of Risk to Safety in Your Daily Life (Coded)')
plt.savefig('Figures/risks.pdf')
plt.show()


df = df.astype({"GlobalRegion": int})
Africa = ['Eastern Africa','Central/Western Africa','North Africa','Southern Africa']
AmericaANZ = ['Latin America & Caribbean','Northern America','Australia and New Zealand']
Asia = ['Central Asia','East Asia','South-eastern Asia','South Asia','Middle East']
Europe = ['Eastern Europe','Northern/Western Europe','Southern Europe']


#Make Africa Scatterpolar chart
Region = Africa
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q4A','Q4B','Q4C','Q4D','Q4E','Q4F','Q4G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)

fig = go.Figure( layout=layout)

for m in range(1,4):
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<4]
    dfm.replace({3: 0, 2: 0.5}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[m],
        fill='toself'
        
    ))

min_value = 0
max_value = 0.8

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

# pio.write_image(fig,'./Figures/Risks_perception_Africa.pdf')
#fig.write_image('./Figures/Risks_perception_Africa.pdf') 
fig.show()


#Make AmericaANZ Scatterpolar chart
Region = AmericaANZ
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q4A','Q4B','Q4C','Q4D','Q4E','Q4F','Q4G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in [5,6,15]:
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<4]
    dfm.replace({3: 0, 2: 0.5}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.8

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

pio.write_image(fig,'./Figures/Risks_perception_AmericaANZ.pdf') 
fig.show()


#Make Asia Scatterpolar chart
Region = Asia
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q4A','Q4B','Q4C','Q4D','Q4E','Q4F','Q4G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in range(7,12):
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<4]
    dfm.replace({3: 0, 2: 0.5}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.8

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

pio.write_image(fig,'./Figures/Risks_perception_Asia.pdf') 
fig.show()


#Make Europe Scatterpolar chart
Region = Europe
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q4A','Q4B','Q4C','Q4D','Q4E','Q4F','Q4G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in [12,13,14]:
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<4]
    dfm.replace({3: 0, 2: 0.5}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.8

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

pio.write_image(fig,'./Figures/Risks_perception_Europe.pdf') 
fig.show()


df_n = df.drop(df[(df['Q5E'] > 4)].index)
df_n = df_n.drop(df_n[(df_n['Q4E'] > 3)].index)


#Make Africa Scatterpolar chart
Region = Africa
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)

fig = go.Figure( layout=layout)

for m in range(1,4):
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<5]
    dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[m],
        fill='toself'
        
    ))

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="center", x=0.35,orientation="h"),
    
)

pio.write_image(fig,'./Figures/Risks_experience_Africa.pdf')

fig.show()


Region = AmericaANZ
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in [5,6,15]:
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<5]
    dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="center", x=0.35,orientation="h"),
    )

pio.write_image(fig,'./Figures/Risks_experience_AmericaANZ.pdf') 
fig.show()


Region = Asia
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in range(7,12):
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<5]
    dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="center", x=0.35,orientation="h"),
    )

pio.write_image(fig,'./Figures/Risks_experience_Asia.pdf') 
fig.show()


#Make Europe Scatterpolar chart
Region = Europe
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in [12,13,14]:
    dfm = df.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    dfm = dfm[dfm<5]
    dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="center", x=0.35,orientation="h"),
    
)

pio.write_image(fig,'./Figures/Risks_experience_Europe.pdf') 
fig.show()


def get_country_acronym(country_name):
    try:
        country = pycountry.countries.search_fuzzy(country_name)[0]
        return country.alpha_3
    except LookupError:
        return None

# Assuming you have a pandas DataFrame named 'df' and the country names are in a column named 'Country'
unique_countries = df['Country'].unique()

country_acronyms = {country: get_country_acronym(country) for country in unique_countries}

#fix the country acronym that is not available in the pycountry package
country_acronyms['Laos']='LAO'
country_acronyms['Bosnia Herzegovina']='BiH'
country_acronyms['Congo Brazzaville'] = 'CG'
country_acronyms['Ivory Coast'] = 'IC'

# Update country names to acronyms
df['Country_short'] = df['Country'].map(country_acronyms)
df.head()


lands = gpd.read_file("./shapefiles/continent.shp")
lands = lands[lands['CONTINENT'] != 'Antarctica']
lands


regions = gpd.read_file("./shapefiles/World_Countries__Generalized_.shp")
regions = regions[regions['COUNTRY'] != 'Antarctica']
regions = regions[['COUNTRY','geometry']]
regions


regions['COUNTRY'] = regions['COUNTRY'].replace('Russian Federation', 'Russia')
regions['COUNTRY'] = regions['COUNTRY'].replace('Bosnia and Herzegovina', 'Bosnia Herzegovina')
regions['COUNTRY'] = regions['COUNTRY'].replace('Congo', 'Congo Brazzaville')
regions['COUNTRY'] = regions['COUNTRY'].replace('Palestinian Territory', 'Palestine')
regions['COUNTRY'] = regions['COUNTRY'].replace('Turkiye', 'Turkey')
regions['COUNTRY'] = regions['COUNTRY'].replace("Côte d'Ivoire", 'Ivory Coast')


#Extracting colums of countru and risks from df
df_1=df[["Country", "Q2_1","GlobalRegion"]].copy()
#then change the name of some countries
df_1['Country'] = df_1['Country'].replace('Hong Kong', 'China');
df_1['Country'] = df_1['Country'].replace('Taiwan', 'China');
df_1['Country'] = df_1['Country'].replace('Kosovo', 'Serbia');


df_1 = df_1.dropna(subset=['Q2_1'])
df_1 = df_1[~df_1['Q2_1'].isin([98, 99])]
df_1


# Create a dictionary to map numerical codes to risk labels
risk_labels = {
    1: 'TRANS1',
    2: 'TRANS2',
    3: 'CRIME',
    4: 'WAR',
    5: 'HEALTH1',
    6: 'HEALTH2',
    7: 'HEALTH3',
    8: 'HEALTH4',
    9: 'FINANCIAL',
    10: 'ECONOMY',
    11: 'POLITICS',
    12: 'TECHNOLOGY',
    13: 'WATER',
    14: 'FOOD1',
    15: 'FOOD2',
    16: 'COOKING',
    17: 'WORK',
    18: 'ENVIRONMENT1',
    19: 'ENVIRONMENT2',
    20: 'ENVIRONMENT3',
    21: 'DROWNING',
    22: 'Other',
    23: 'Nothing',
}

# Replace numerical codes with risk labels using the map() method
df_1['risk_category'] = df_1['Q2_1'].map(risk_labels)
#drop NaN raws
# df_1 = df_1.dropna(subset=['risk_category'])
df_1


globalregion_labels = {
    1: 'Eastern Africa',
    2: 'Central/Western Africa',
    3: 'North Africa',
    4: 'Southern Africa',
    5: 'Latin America & Caribbean',
    6: 'Northern America',
    7: 'Central Asia',
    8: 'East Asia',
    9: 'South-eastern Asia',
    10: 'South Asia',
    11: 'Middle East',
    12: 'Eastern Europe',
    13: 'Northern/Western Europe',
    14: 'Southern Europe',
    15: 'Australia and New Zealand',
}

# Replace numerical codes with risk labels using the map() method
df_1['globalregion_category'] = df_1['GlobalRegion'].map(globalregion_labels)
#drop NaN raws
# df_1 = df_1.dropna(subset=['globalregion_category'])
df_1


# Merge world data base with Risk Poll data base
merged = regions.merge(df_1, left_on='COUNTRY', right_on='Country')
#merged = merged.drop(merged[merged['Q2_1'] > 25].index)
merged


merged_first = merged.groupby('Country').first().reset_index()
merged_first = merged_first.drop(['Q2_1', 'risk_category','GlobalRegion','COUNTRY','GlobalRegion'], axis=1)


# Set the size of the plot
fig, ax = plt.subplots(figsize=(8, 8), dpi=150)

# Plot the base map in grey color
regions.plot(ax=ax,facecolor='lightgrey', edgecolor='white', linewidth=0.5)

# Plot the merged_first map on top of the base map
merged_first.plot(ax=plt.gca(), column='Country', color='green', edgecolor='white')
# merged_first.plot(ax=plt.gca(), column='COUNTRY', cmap='Blues', edgecolor='white', legend=True, legend_kwds={'label': 'globalregion_category'})


# add markers
for idx, row in merged_first.iterrows():
    plt.scatter(row.geometry.centroid.x, row.geometry.centroid.y, s=5, color='red')
    plt.annotate(row['Country'], xy=(row.geometry.centroid.x, row.geometry.centroid.y), xytext=(5, 5), textcoords='offset points', fontsize=4, color='black')

# show the plot
plt.show()

/Users/mkie815/opt/anaconda3/envs/ml/lib/python3.9/site-packages/geopandas/plotting.py:656: UserWarning:

Only specify one of 'column' or 'color'. Using 'color'.


participants = df.dropna(subset=['Q2_1'])
participants = df.groupby('Country').size().reset_index(name = 'Frequency_country')

merged_transport1 = merged[merged['Q2_1'].isin([1])]
merged_transport1 = merged_transport1.groupby('Country').size().reset_index(name = 'Frequency_risks')

merged_transport1 = pd.merge(merged_transport1, participants, on='Country')
#risk percentage
merged_transport1['Average_risks_percent'] = (merged_transport1['Frequency_risks'] / merged_transport1['Frequency_country']) * 100
# merge with the geomatrical data
merged_transport1 = merged_first.merge(merged_transport1, left_on='Country', right_on='Country')


import matplotlib as mpl
cmap = mpl.cm.YlOrBr

# Set the size of the plot
fig, ax = plt.subplots(figsize=(8, 8), dpi=150)

# Plot the base map in grey color
regions.plot(ax=ax, facecolor='lightgrey', edgecolor='white', linewidth=0.5)

# Remove grids
ax.grid(False)

# Risk frequency
risk_plot = merged_transport1.plot(ax=ax, column='Average_risks_percent', cmap=cmap, legend=False)

# Create a colorbar within the main graph and make it horizontal
norm = plt.Normalize(merged_transport1['Average_risks_percent'].min(), merged_transport1['Average_risks_percent'].max())
cbar_ax = fig.add_axes([0.25, 0.35, 0.5, 0.02])  # Adjust the position and size of the colorbar
cbar = mcb.ColorbarBase(cbar_ax, cmap=cmap, norm=norm, orientation='horizontal')
cbar.ax.set_xlabel('Average transport risks (%)',fontsize = 6)
cbar.ax.tick_params(labelsize=6)

# Add markers and store annotations in a list
texts = []
for idx, row in merged_first.iterrows():
    ax.scatter(row.geometry.centroid.x, row.geometry.centroid.y, s=3, color='red')
    if (row['globalregion_category'] != 'Eastern Europe') &  (row['globalregion_category'] != 'Southern Europe') :
    
        texts.append(ax.annotate(row['Country'], xy=(row.geometry.centroid.x, row.geometry.centroid.y), xytext=(5, 5), textcoords='offset points', fontsize=4, color='black'))

# Optimize annotation positions to minimize overlaps
adjust_text(texts, ax=ax)

# Hide the x-axis and y-axis ticks and labels
ax.set_xticks([])
ax.set_yticks([])
plt.tight_layout()
plt.savefig('./Figures/transport_map.pdf')
# Show the plo
plt.show()

/var/folders/c9/h5rgb9zs5p5d8dqmbtbjjq_h0000gr/T/ipykernel_7850/2265349828.py:37: UserWarning:

This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.


# Experience traffic issues
df_traffic= df[~df['Q5E'].isin([98, 99])].copy()

df_traffic['Q5E'].replace({3: 1, 1: 0.75, 2: 0.5, 4: 0}, inplace=True)

x= df_traffic.groupby('Country').mean()['Q5E']
# merge with the geomatrical data
merged_transport_experience = merged_first.merge(x, left_on='Country', right_on='Country')
merged_transport_experience.head()


# Set the size of the plot
fig, ax = plt.subplots(figsize=(8, 4), dpi=150)

lands.plot(ax=ax, cmap='Pastel1', edgecolor='white', linewidth=0.5)
# Plot the base map in grey color
regions.plot(ax=ax,facecolor='none', edgecolor='white', linewidth=0.5)

# Plot the merged_first map on top of the base map
#merged_transport1.plot(ax=plt.gca(), column='Average_risks_percent', cmap='RdYlBu', edgecolor='white')
merged_transport_experience['Experience_formated'] = merged_transport_experience['Q5E'].apply(lambda x: "{:.2f}".format(x))
merged_transport_experience.loc[:,'Rank'] = merged_transport_experience['Q5E'].rank(ascending=False)
df_transport_ranked = merged_transport_experience[merged_transport_experience['Rank']<21].copy()

# add markers
for idx, row in df_transport_ranked.iterrows():
    plt.scatter(row.geometry.centroid.x, row.geometry.centroid.y, s=3, color='red')
    plt.annotate(f"{int(row['Rank'])}", xy=(row.geometry.centroid.x, row.geometry.centroid.y), xytext=(5, 5), textcoords='offset points', fontsize=4, color='black')
    
# Create a list of strings for the legend
df_transport_ranked = df_transport_ranked.sort_values(by='Rank')
legend_labels = [f'{int(row["Rank"])}: {row["Country"]} ({row["Experience_formated"]})' for _, row in df_transport_ranked.iterrows()]
plt.legend(legend_labels, loc='upper left', fontsize=6,ncol=5)

# Hide the x-axis and y-axis ticks and labels
ax.set_xticks([])
ax.set_yticks([])
plt.savefig('./Figures/transport_experience_rank_map.pdf')
# show the plot
plt.show();


merged_crime = merged[merged['Q2_1'].isin([3])]

merged_crime = merged_crime.groupby(['Country', 'risk_category']).size().reset_index(name='Frequency_risks')

merged_crime = pd.merge(merged_crime, participants, on='Country')
#risk percentage
merged_crime['Average_risks_percent'] = (merged_crime['Frequency_risks'] / merged_crime['Frequency_country']) * 100
# merge with the geomatrical data
merged_crime = merged_first.merge(merged_crime, left_on='Country', right_on='Country')
#merged_crime


cmap = mpl.cm.YlOrBr
# Set the size of the plot
fig, ax = plt.subplots(figsize=(8, 8), dpi=150)

# Plot the base map in grey color
regions.plot(ax=ax, facecolor='lightgrey', edgecolor='white', linewidth=0.5)

# Remove grids
ax.grid(False)

# Risk frequency
risk_plot = merged_crime.plot(ax=ax, column='Average_risks_percent', cmap=cmap, legend=False)

# Create a colorbar within the main graph and make it horizontal
norm = plt.Normalize(merged_crime['Average_risks_percent'].min(), merged_crime['Average_risks_percent'].max())
cbar_ax = fig.add_axes([0.25, 0.35, 0.5, 0.02])  # Adjust the position and size of the colorbar
cbar = mcb.ColorbarBase(cbar_ax, cmap=cmap, norm=norm, orientation='horizontal')
cbar.ax.set_xlabel('Average crime risks (%)',fontsize = 6)
cbar.ax.tick_params(labelsize=6)

# Add markers and store annotations in a list
texts = []
for idx, row in merged_first.iterrows():
    ax.scatter(row.geometry.centroid.x, row.geometry.centroid.y, s=3, color='red')
    if (row['globalregion_category'] != 'Eastern Europe') &  (row['globalregion_category'] != 'Southern Europe') :
        texts.append(ax.annotate(row['Country'], xy=(row.geometry.centroid.x, row.geometry.centroid.y), xytext=(5, 5), textcoords='offset points', fontsize=4, color='black'))

# Optimize annotation positions to minimize overlaps
adjust_text(texts, ax=ax)

# Hide the x-axis and y-axis ticks and labels
ax.set_xticks([])
ax.set_yticks([])
#plt.tight_layout()
plt.savefig('./Figures/crime_map.pdf')
# Show the plot
plt.show()


# Experience traffic issues
df_crime= df[~df['Q5C'].isin([98, 99])].copy()

df_crime['Q5C'].replace({3: 1, 1: 0.75, 2: 0.5, 4: 0}, inplace=True)

x= df_crime.groupby('Country').mean()['Q5C']
# merge with the geomatrical data
merged_crime_experience = merged_first.merge(x, left_on='Country', right_on='Country')
merged_crime_experience.head()


# Set the size of the plot
fig, ax = plt.subplots(figsize=(8, 8), dpi=150)

lands.plot(ax=ax, cmap='Pastel1', edgecolor='white', linewidth=0.5)
# Plot the base map in grey color
regions.plot(ax=ax,facecolor='none', edgecolor='white', linewidth=0.5)

# Plot the merged_first map on top of the base map
merged_crime_experience['Experience_formated'] = merged_crime_experience['Q5C'].apply(lambda x: "{:.2f}".format(x))
merged_crime_experience.loc[:,'Rank'] = merged_crime_experience['Q5C'].rank(ascending=False)
df_crime_ranked = merged_crime_experience[merged_crime_experience['Rank']<21].copy()

# add markers
for idx, row in df_crime_ranked.iterrows():
    plt.scatter(row.geometry.centroid.x, row.geometry.centroid.y, s=3, color='red')
    plt.annotate(f"{int(row['Rank'])}", xy=(row.geometry.centroid.x, row.geometry.centroid.y), xytext=(5, 5), textcoords='offset points', fontsize=4, color='black')
    
# Create a list of strings for the legend
df_crime_ranked = df_crime_ranked.sort_values(by='Rank')
legend_labels = [f'{int(row["Rank"])}: {row["Country"]} ({row["Experience_formated"]})' for _, row in df_crime_ranked.iterrows()]
plt.legend(legend_labels,  loc='upper left', fontsize=6,ncol=5)

# Hide the x-axis and y-axis ticks and labels
ax.set_xticks([])
ax.set_yticks([])
plt.savefig('./Figures/crime_experience_rank_map.pdf')
# show the plot
plt.show();


# worrylevel_traffic_mean
df_new_1= df[~df['Q4E'].isin([98, 99])]
expected_perception_traffic = np.mean(df_new_1['Q4E'])
# worrylevel_traffic_SD
std_perception_traffic = np.std(df_new_1['Q4E'])
# worrylevel_crime_mean
df_new_2= df[~df['Q4C'].isin([98, 99])]
expected_perception_crime = np.mean(df_new_2['Q4C'])
# # worrylevel_crime_SD
std_perception_crime = np.std(df_new_2['Q4C'])
# worrylevel_food_mean
df_new_3= df[~df['Q4A'].isin([98, 99])]
expected_perception_food = np.mean(df_new_3['Q4A'])
# # worrylevel_food_SD
std_perception_food = np.std(df_new_3['Q4A'])
# worrylevel_water_mean
df_new_4= df[~df['Q4B'].isin([98, 99])]
expected_perception_water = np.mean(df_new_4['Q4B'])
# # worrylevel_water_SD
std_perception_water = np.std(df_new_4['Q4B'])
# worrylevel_weather_mean
df_new_5= df[~df['Q4D'].isin([98, 99])]
expected_perception_weather = np.mean(df_new_5['Q4D'])
# # worrylevel_weather_SD
std_perception_weather = np.std(df_new_5['Q4D'])
# worrylevel_mental_health_mean
df_new_6= df[~df['Q4F'].isin([98, 99])]
expected_perception_health = np.mean(df_new_6['Q4F'])
# # worrylevel_mental_health_SD
std_perception_health = np.std(df_new_6['Q4F'])
# worrylevel_work_you_do_mean
df_new_7= df[~df['Q4G'].isin([98, 99])]
expected_perception_work = np.mean(df_new_7['Q4G'])
# # worrylevel_work_you_do_SD
std_perception_work = np.std(df_new_7['Q4G'])


mean = pd.Series([1.8510530749789385, 1.9285030299043149, 2.182904564315353, 2.3037420023835327, 1.9133533574301278, 2.256654954631144, 2.3610715228292425])
sd= pd.Series([0.7829518465178122, 0.82915454740197, 0.7773544875867079, 0.7903374432342158, 0.8041756714545396, 0.8066594199051904, 0.7573198790556099])
df_mean_sd = pd.concat ([mean,sd], axis=1)
df_mean_sd.columns = ['Mean','SD']
df_mean_sd.index = ['Traffic','Crime','Food','Water', 'Environment 2', 'Health 4', 'Work']
df_mean_sd.index.name = 'Risks'
df_mean_sd


index=df.loc[:, ['GlobalRegion','Country', 'Q4A', 'Q4B','Q4C','Q4D', 'Q4E','Q4F','Q4G','Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G','Q20']]
index.head()


cols_to_replace = ['Q4A', 'Q4B','Q4C','Q4D', 'Q4E','Q4F','Q4G','Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G','Q20']
replace_dict = {98: np.nan, 99: np.nan}
index[cols_to_replace] = index[cols_to_replace].replace(replace_dict)
index.head()


#removing the raws if all the columns have missing values, 98, 99 
index.dropna(subset=['Q4A', 'Q4B','Q4C','Q4D', 'Q4E','Q4F','Q4G','Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G','Q20'], thresh=7, inplace=True)
index.head()


# replace the values 2 with 0.5 and 3 with 0
index.loc[:, ['Q4A', 'Q4B','Q4C','Q4D', 'Q4E','Q4F','Q4G']] = index.loc[:, ['Q4A', 'Q4B','Q4C','Q4D', 'Q4E','Q4F','Q4G']].replace({2: 0.5, 3: 0})
index.loc[:, ['Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G']] = index.loc[:, ['Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G']].replace({3: 1, 1: 0.75, 2: 0.5, 4: 0})
index.loc[:, ['Q20']] = index.loc[:, ['Q20']].replace({3: 0.5, 2: 0})
index.head()


# define a function to compute the average of the non-ignored values in each row for perception score (worry) and experience score)
def compute_average(row):
    valid_values = [value for value in row if not pd.isna(value)]
    if len(valid_values) > 0:
        return np.mean(valid_values)
    else:
        return np.nan

# create a new column with the computed average
index['overall_perception_score'] = index.loc[:, ['Q4A', 'Q4B','Q4C','Q4D', 'Q4E','Q4F','Q4G']].apply(compute_average, axis=1)
index['overall_experience_score'] = index.loc[:, ['Q5A','Q5B','Q5C','Q5D','Q5E','Q5F','Q5G']].apply(compute_average, axis=1)
index.head()


#rename the columns
index = index.rename(columns={'Q4A': 'Food_worry', 'Q4B': 'water_worry','Q4C': 'crime_worry', 'Q4D': 'weather_worry', 'Q4E': 'traffic_worry','Q4F': 'mental_health_worry','Q4G': 'work_worry','Q5A': 'Food_experience', 'Q5B': 'water_experience','Q5C': 'crime_experience', 'Q5D': 'weather_experience', 'Q5E': 'traffic_experience','Q5F': 'mental_health_experience','Q5G': 'work_experience', 'Q20': 'readiness_score'})
index.head()


#calculating Individual impact for each country from each risk
index['risk_impact_food'] = index['Food_worry'] * index['Food_experience']
index['risk_impact_water'] = index['water_worry'] * index['water_experience']
index['risk_impact_crime'] = index['crime_worry'] * index['crime_experience']
index['risk_impact_weather'] = index['weather_worry'] * index['weather_experience']
index['risk_impact_traffic'] = index['traffic_worry'] * index['traffic_experience']
index['risk_impact_mental_health'] = index['mental_health_worry'] * index['mental_health_experience']
index['risk_impact_work'] = index['work_worry'] * index['work_experience']
index['overall_risk_impact_score'] = index['overall_perception_score'] * index['overall_experience_score']
index.head()


#Make Africa Scatterpolar chart
Region = Africa
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['risk_impact_food','risk_impact_water','risk_impact_crime','risk_impact_weather','risk_impact_traffic','risk_impact_mental_health','risk_impact_work']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)

fig = go.Figure( layout=layout)

for m in range(1,4):
    dfm = index.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
#     dfm = dfm[dfm<5]
#     dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[m],
        fill='toself'
        
    ))

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

pio.write_image(fig,'./Figures/Risks_impact_Africa.pdf')

fig.show()


Region = AmericaANZ
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['risk_impact_food','risk_impact_water','risk_impact_crime','risk_impact_weather','risk_impact_traffic','risk_impact_mental_health','risk_impact_work']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in [5,6,15]:
    dfm = index.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
#     dfm = dfm[dfm<5]
#     dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

pio.write_image(fig,'./Figures/Risks_impact_AmericaANZ.pdf') 
fig.show()


Region = Asia
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['risk_impact_food','risk_impact_water','risk_impact_crime','risk_impact_weather','risk_impact_traffic','risk_impact_mental_health','risk_impact_work']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in range(7,12):
    dfm = index.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    # dfm = dfm[dfm<5]
    # dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

pio.write_image(fig,'./Figures/Risks_impact_Asia.pdf') 
fig.show()


Region = Europe
Categories = ['Food','Water','Crime','Serious Weather','Traffic','Mental health','Work accident']
Col = ['risk_impact_food','risk_impact_water','risk_impact_crime','risk_impact_weather','risk_impact_traffic','risk_impact_mental_health','risk_impact_work']
layout = go.Layout(
    autosize=True,  # Enable autosizing
    width=500,  # Adjust the width of the plot
    height=400,  # Adjust the height of the plot
    margin=dict(l=5, r=0, t=15, b=2)  # Set the left, right, top, and bottom margins to zero
)
fig = go.Figure(layout=layout)
count0=0
for m in [12,13,14]:
    dfm = index.loc[df.GlobalRegion==m,Col]
    dfm.dropna(inplace=True)
    # dfm = dfm[dfm<5]
    # dfm.replace({4: 0, 3: 1, 2: 0.5, 1:0.75}, inplace=True)
    value= dfm.mean()

    fig.add_trace(go.Scatterpolar(
        theta=Categories,
        r=value,
        name=Region[count0],
        fill='toself'
        
    ))
    count0+=1

min_value = 0
max_value = 0.35

fig.update_layout(
    polar=dict(
        radialaxis=dict(
            range=[min_value, max_value]  # Set the minimum and maximum values for the r-axis
        )
    ),
    legend=dict(yanchor="top", y=-0.15, xanchor="left", x=0.4),
    legend_orientation="v"
)

pio.write_image(fig,'./Figures/Risks_impact_Europe.pdf') 
fig.show()


mean_scores = index.groupby('Country').agg({'overall_perception_score': 'mean', 'overall_experience_score': 'mean', 'overall_risk_impact_score': 'mean', 'readiness_score': 'mean','Food_worry': 'mean', 'water_worry': 'mean', 'crime_worry': 'mean', 'weather_worry': 'mean', 'traffic_worry': 'mean', 'mental_health_worry': 'mean', 'work_worry': 'mean', 'Food_experience': 'mean', 'water_experience': 'mean', 'crime_experience': 'mean', 'weather_experience': 'mean', 'traffic_experience': 'mean', 'mental_health_experience': 'mean', 'work_experience': 'mean', 'risk_impact_food': 'mean','risk_impact_water': 'mean','risk_impact_crime': 'mean','risk_impact_weather': 'mean','risk_impact_traffic': 'mean','risk_impact_mental_health': 'mean','risk_impact_work': 'mean'}).reset_index()
mean_scores = mean_scores.rename(columns={'overall_perception_score': 'perception_index', 'overall_experience_score': 'experience_index', 'overall_risk_impact_score': 'risk_impact_index','readiness_score': 'readiness_index', 'Food_worry': 'Perception_index_food', 'water_worry': 'Perception_index_water', 'crime_worry': 'Perception_index_crime', 'weather_worry': 'Perception_index_weather', 'traffic_worry': 'Perception_index_traffic',  'mental_health_worry': 'Perception_index_mental_health', 'work_worry': 'Perception_index_work','Food_experience': 'experience_index_food', 'water_experience': 'experience_index_water','crime_experience': 'experience_index_crime', 'weather_experience': 'experience_index_weather', 'traffic_experience': 'experience_index_traffic','mental_health_experience': 'experience_index_mental_health', 'work_experience': 'experience_index_work','risk_impact_food': 'risk_impact_index_food','risk_impact_water': 'risk_impact_index_water','risk_impact_crime': 'risk_impact_index_crime','risk_impact_weather': 'risk_impact_index_weather','risk_impact_traffic': 'risk_impact_index_traffic','risk_impact_mental_health': 'risk_impact_index_mental_health','risk_impact_work': 'risk_impact_index_work' })
mean_scores.head()


mean_scores.isnull().any()

Country                            False
perception_index                   False
experience_index                   False
risk_impact_index                  False
readiness_index                    False
Perception_index_food              False
Perception_index_water             False
Perception_index_crime             False
Perception_index_weather           False
Perception_index_traffic           False
Perception_index_mental_health     False
Perception_index_work              False
experience_index_food              False
experience_index_water             False
experience_index_crime             False
experience_index_weather           False
experience_index_traffic           False
experience_index_mental_health     False
experience_index_work              False
risk_impact_index_food             False
risk_impact_index_water            False
risk_impact_index_crime            False
risk_impact_index_weather          False
risk_impact_index_traffic          False
risk_impact_index_mental_health    False
risk_impact_index_work             False
dtype: bool


zeros_present = (mean_scores == 0).any()
zeros_present

Country                            False
perception_index                   False
experience_index                   False
risk_impact_index                  False
readiness_index                    False
Perception_index_food              False
Perception_index_water             False
Perception_index_crime             False
Perception_index_weather           False
Perception_index_traffic           False
Perception_index_mental_health     False
Perception_index_work              False
experience_index_food              False
experience_index_water             False
experience_index_crime             False
experience_index_weather           False
experience_index_traffic           False
experience_index_mental_health     False
experience_index_work              False
risk_impact_index_food             False
risk_impact_index_water            False
risk_impact_index_crime            False
risk_impact_index_weather          False
risk_impact_index_traffic          False
risk_impact_index_mental_health    False
risk_impact_index_work             False
dtype: bool


# keep only the 'Country' and 'Global Region' columns from worry
index_region = index[['Country', 'GlobalRegion']]

# remove any duplicate rows for each country
index_region = index_region.drop_duplicates(subset='Country', keep='first')

# reset the index of the dataframe
index_region = index_region.reset_index(drop=True)

# Define a dictionary to map the numbers to the region names
region_map = {
    1: 'Eastern Africa',
    2: 'Central/Western Africa',
    3: 'North Africa',
    4: 'Southern Africa',
    5: 'Latin America & Caribbean',
    6: 'Northern America',
    7: 'Central Asia',
    8: 'East Asia',
    9: 'South-eastern Asia',
    10: 'South Asia',
    11: 'Middle East',
    12: 'Eastern Europe',
    13: 'Northern/Western Europe',
    14: 'Southern Europe',
    15: 'Australia and New Zealand'
}

# Use the replace method to replace the numbers with the region names
index_region['GlobalRegion'] = index_region['GlobalRegion'].replace(region_map)
index_region.head()


index_scores = mean_scores.merge(index_region, on='Country')
index_scores.head()


index_scores.isna().sum()

Country                            0
perception_index                   0
experience_index                   0
risk_impact_index                  0
readiness_index                    0
Perception_index_food              0
Perception_index_water             0
Perception_index_crime             0
Perception_index_weather           0
Perception_index_traffic           0
Perception_index_mental_health     0
Perception_index_work              0
experience_index_food              0
experience_index_water             0
experience_index_crime             0
experience_index_weather           0
experience_index_traffic           0
experience_index_mental_health     0
experience_index_work              0
risk_impact_index_food             0
risk_impact_index_water            0
risk_impact_index_crime            0
risk_impact_index_weather          0
risk_impact_index_traffic          0
risk_impact_index_mental_health    0
risk_impact_index_work             0
GlobalRegion                       0
dtype: int64


#overall resilience index
index_scores['resilience_index'] = index_scores['risk_impact_index'] * index_scores['readiness_index']
index_scores.head()


df_traffic = pd.read_csv("shapefiles/traffic.csv")
df_traffic.head()


# remove the part in brackets and convert the remaining string to integer
df_traffic['Estimated number of road traffic deaths'] = df_traffic['Estimated number of road traffic deaths'].str.replace(r'\[.*\]', '').astype(int)
df_traffic['Estimated road traffic death rate (per 100 000 population)'] = df_traffic['Estimated road traffic death rate (per 100 000 population)'].str.replace(r'\[.*\]', '').astype(float)
df_traffic= df_traffic[df_traffic['Year'].isin([2019])]

/var/folders/cf/v8stgyrj37b118_clh0v_7wm0000gn/T/ipykernel_12986/1495670512.py:2: FutureWarning:

The default value of regex will change from True to False in a future version.

/var/folders/cf/v8stgyrj37b118_clh0v_7wm0000gn/T/ipykernel_12986/1495670512.py:3: FutureWarning:

The default value of regex will change from True to False in a future version.


#rename the countries in statistical traffic data as Poll data survey
df_traffic['Country'] = df_traffic['Country'].replace('Bolivia (Plurinational State of)', 'Bolivia')
df_traffic['Country'] = df_traffic['Country'].replace('Bosnia and Herzegovina', 'Bosnia Herzegovina')
df_traffic['Country'] = df_traffic['Country'].replace('Congo', 'Congo Brazzaville')
df_traffic['Country'] = df_traffic['Country'].replace('Czechia', 'Czech Republic')
df_traffic['Country'] = df_traffic['Country'].replace('Iran (Islamic Republic of)', 'Iran')
df_traffic['Country'] = df_traffic['Country'].replace("Côte d'Ivoire", 'Ivory Coast')
df_traffic['Country'] = df_traffic['Country'].replace("Lao People's Democratic Republic", 'Laos')
df_traffic['Country'] = df_traffic['Country'].replace('Republic of Moldova', 'Moldova')
df_traffic['Country'] = df_traffic['Country'].replace('Russian Federation', 'Russia')
df_traffic['Country'] = df_traffic['Country'].replace('Republic of Korea', 'South Korea')
df_traffic['Country'] = df_traffic['Country'].replace('United Republic of Tanzania', 'Tanzania')
df_traffic['Country'] = df_traffic['Country'].replace('Türkiye', 'Turkey')
df_traffic['Country'] = df_traffic['Country'].replace('United Kingdom of Great Britain and Northern Ireland', 'United Kingdom')
df_traffic['Country'] = df_traffic['Country'].replace('United States of America', 'United States')
df_traffic['Country'] = df_traffic['Country'].replace('Venezuela (Bolivarian Republic of)', 'Venezuela')
df_traffic['Country'] = df_traffic['Country'].replace('Viet Nam', 'Vietnam')


merged_traffic = df_traffic.merge(index_scores, on='Country', how='inner')
merged_traffic.head()


# Define the region categories
Africa = ['Eastern Africa', 'Central/Western Africa', 'North Africa', 'Southern Africa']
North_America = ['Northern America']
South_America = ['Latin America & Caribbean']
Oceania = ['Australia and New Zealand'] 
Asia = ['Central Asia', 'East Asia', 'South-eastern Asia', 'South Asia', 'Middle East']
Europe = ['Eastern Europe', 'Northern/Western Europe', 'Southern Europe']

# Function to map country to region category
def map_region_category(country):
    if country in Africa:
        return 'Africa'
    elif country in North_America:
        return 'North America'
    elif country in South_America:
        return 'South America'
    elif country in Oceania:
        return 'Oceania'
    elif country in Asia:
        return 'Asia'
    elif country in Europe:
        return 'Europe'
    else:
        return 'Other'


# Create a custom color palette for each region category
color_palette = {
    'Africa': 'orange',
    'North America': 'green',
    'South America': 'Pink',
    'Asia': 'blue',
    'Europe': 'red',
    'Oceania': 'Yellow',
}


# List of countries to label
countries_to_label = ['United Kingdom','Malaysia','Jordan','South Korea','Albania','Mongolia','Macedonia','Zambia','Nigeria','Benin','Burkina Faso','Togo','Cameroon','Mali','Finland','Egypt','Latvia','Kazakhstan','Paraguay','United States','Germany','Spain','Afganistan','Thailan','Algeria','Peru','Indonesia','Laos','Serbia','Ukraine','Colombia','Uzbekistan','Guinea','Mali','Dominican Republic','Venezuela','Saudi Arabia','Namibia','Thailand','El Salvador','Paraguay','New Zealand','Australia','United States','China','Mexico','Philippines','Israel','Slovakia','Vietnam','Iraq','Nigeria','Ghana','Namibia','Sri Lanka','Iran','Cambodia','Kenya','Ivory Coast','Brazil','South Africa','Mozambique','Togo','China','Mexico','Laos','Sweden','France','Ireland','Japan','Mauritius']
#countries_to_label = merged_traffic.Country.unique()

# Create a scatter plot
plt.subplots(figsize=(8, 8), dpi=200)

# Apply the function to create the 'RegionCategory' column
merged_traffic['RegionCategory'] = merged_traffic['GlobalRegion'].apply(map_region_category)

sns.scatterplot(
    data=merged_traffic,
    x='Perception_index_traffic',
    y='experience_index_traffic',
    hue='RegionCategory',
    palette=color_palette
)

# Add country names as labels to the points
for i, row in merged_traffic.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['Perception_index_traffic'], row['experience_index_traffic'], row['Country'], fontsize=12)


# Add a regression line
sns.regplot(
    data=merged_traffic,
    x='Perception_index_traffic',
    y='experience_index_traffic',
    scatter=False,  # Set to False to avoid plotting data points again
    color='black',  # Set the color of the regression line
    line_kws={"linewidth": 1}  # Set the line width of the regression line
)

# Set x and y axis labels
plt.xlabel('Traffic Risk Perception index', fontsize=16)
plt.ylabel('Traffic Risk Experience index', fontsize=16)

#plt.legend(handles=legend_handles, loc='upper left', fontsize=14)  #bbox_to_anchor=(1.05, 1),
plt.legend(loc='upper left', fontsize=14)  #bbox_to_anchor=(1.05, 1),

plt.savefig('./Figures/traffic_perception_experience.pdf',bbox_inches='tight')

# Display the plot
plt.show()


# List of countries to label
countries_to_label = ['Switzerland','United Kingdom','Uzbekistan','Guinea','Mali','Dominican Republic','Zimbabwe','Venezuela','Saudi Arabia','Namibia','Sierra Leone','Thailand','El Salvador','Paraguay','New Zealand','Australia','United States','China','Mexico','Philippines','Israel','Slovakia','Poland','Vietnam','Iraq','Nigeria','Ghana','Namibia','Sri Lanka','Iran','Cambodia','Kenya','Ivory Coast','Brazil','South Africa','Mozambique','Togo','China','Mexico','Laos']
#countries_to_label = merged_traffic.Country.unique()

# Create a scatter plot
plt.subplots(figsize=(8, 8), dpi=200)

# Apply the function to create the 'RegionCategory' column
merged_traffic['RegionCategory'] = merged_traffic['GlobalRegion'].apply(map_region_category)

sns.scatterplot(
    data=merged_traffic,
    x='risk_impact_index_traffic',
    y='Estimated road traffic death rate (per 100 000 population)',
    hue='RegionCategory',
    palette=color_palette
)

# Add country names as labels to the points
for i, row in merged_traffic.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['risk_impact_index_traffic'], row['Estimated road traffic death rate (per 100 000 population)'], row['Country'], fontsize=12)


# Add a regression line
sns.regplot(
    data=merged_traffic,
    x='risk_impact_index_traffic',
    y='Estimated road traffic death rate (per 100 000 population)',
    scatter=False,  # Set to False to avoid plotting data points again
    color='black',  # Set the color of the regression line
    line_kws={"linewidth": 1}  # Set the line width of the regression line
)

# Set x and y axis labels
plt.xlabel('Traffic Risk Impact index', fontsize=16)
plt.ylabel('Estimated road traffic death rate (per 100,000 capita)', fontsize=16)

#plt.legend(handles=legend_handles, loc='upper left', fontsize=14)  #bbox_to_anchor=(1.05, 1),
plt.legend(loc='upper left', fontsize=14)  #bbox_to_anchor=(1.05, 1),

plt.savefig('./Figures/Risks_impact_vs_actual_traffic.pdf',bbox_inches='tight')

# Display the plot
plt.show()


df_crime = pd.read_csv("./shapefiles/crime.csv")
df_crime = df_crime.rename(columns={'country': 'Country'}) 
df_crime.head()


#rename the countries in statistical traffic data as Poll data survey
df_crime['Country'] = df_crime['Country'].replace('Bosnia And Herzegovina', 'Bosnia Herzegovina')
df_crime['Country'] = df_crime['Country'].replace('Papua New Guinea', 'Guinea')


merged_crime = df_crime.merge(index_scores, on='Country', how='inner')
merged_crime.head()


# Create a scatter plot
plt.subplots(figsize=(8, 8), dpi=200)

# Apply the function to create the 'RegionCategory' column
merged_crime['RegionCategory'] = merged_crime['GlobalRegion'].apply(map_region_category)

sns.scatterplot(
    data=merged_crime,
    x='Perception_index_crime',
    y='experience_index_crime',
    hue='RegionCategory',
    palette=color_palette
)

# Add a regression line
sns.regplot(
    data=merged_crime,
    x='Perception_index_crime',
    y='experience_index_crime',
    scatter=False,  # Set to False to avoid plotting data points again
    color='black',  # Set the color of the regression line
    line_kws={"linewidth": 1}  # Set the line width of the regression line
)

countries_to_label = ['United Kingdom','United Arab Emirates','Guinea','Japan','Zimbabwe','Venezuela','Namibia','Thailand','France','New Zealand','United States','China','Mexico','Israel','Afghanistan','South Africa','Bangladesh','Zambia','Canada','Peru','China','Nepal','Iraq','India','Costa Rica','Mongolia','Hungary','El Salvador','Kyrgyzstan','Ghana','France','Finland','Sweden','Denmark','Norway','Israel','India','Afganistan','Mauritius','South Korea','Cyprus','Chile','Brazil','Tanzania','Slovakia','Ukraine','Uzbekistan','Uganda','Venezuela','Colombia']

#countries_to_label = merged_crime.Country.unique()
# Add country names as labels to the points
for i, row in merged_crime.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['Perception_index_crime'], row['experience_index_crime'], row['Country'],  fontsize=12)

# Set x and y axis labels
plt.xlabel('Crime Perception index',fontsize=16)
plt.ylabel('Crime Experience index',fontsize=16)

# Move legend outside plot area
plt.legend(loc='upper left', fontsize=14)

# Display the plot

plt.savefig('./Figures/crime_experience_perception.pdf',bbox_inches='tight')
plt.show();


# List of countries to label
countries_to_label = ['United Kingdom','United Arab Emirates','Guinea','Japan','Dominican Republic','Zimbabwe','Venezuela','Saudi Arabia','Namibia','Singapore','Thailand','France','New Zealand','United States','China','Mexico','Honduras','Israel','Afghanistan','South Africa','Bangladesh','Zambia','Canada','Germany','Algeria','Morocco','Bolivia','Peru','China','Nepal','Brazil','Cameroon','Iraq','India','Costa Rica','Mongolia','Hungary','Spain','El Salvador','Kyrgyzstan','Ghana']
#countries_to_label = merged_traffic.Country.unique()

# Create a scatter plot
plt.subplots(figsize=(8, 8), dpi=150)

# Apply the function to create the 'RegionCategory' column
merged_traffic['RegionCategory'] = merged_crime['GlobalRegion'].apply(map_region_category)

g = sns.scatterplot(
    data=merged_crime,
    x='risk_impact_index_crime',
    y='crimeIndex',
    hue=merged_traffic['RegionCategory'],
    palette=color_palette
)

# Add a regression line
sns.regplot(
    data=merged_crime,
    x='risk_impact_index_crime',
    y='crimeIndex',
    scatter=False,  # Set to False to avoid plotting data points again
    color='black',  # Set the color of the regression line
    line_kws={"linewidth": 1}  # Set the line width of the regression line
)

# Add country names as labels to the points
for i, row in merged_crime.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['risk_impact_index_crime'], row['crimeIndex'], row['Country'], fontsize=12)

# Set x and y axis labels
plt.xlabel('Crime Risks Impact index', fontsize = 16)
plt.ylabel('Crime index', fontsize=16)
g.get_legend().remove()

# Show all the categories in the legend
#legend_labels = ['Africa', 'North America','South America','Asia', 'Europe','Oceana']
#legend_handles = []
#for label in legend_labels:
#    handle = plt.Line2D([], [], color=color_palette[label], marker='o', markersize=5, label=label)
#    legend_handles.append(handle)

#plt.legend(handles=legend_handles, bbox_to_anchor=(1.05, 1), loc='upper left', fontsize=16)
#plt.legend(  loc='lower right', fontsize=14)


plt.savefig('./Figures/Risks_impact_vs_actual_crime.pdf',bbox_inches='tight')

# Display the plot
plt.show()


# Create a scatter plot
plt.subplots(figsize=(8, 8), dpi=200)

# Apply the function to create the 'RegionCategory' column
index_scores['RegionCategory'] = index_scores['GlobalRegion'].apply(map_region_category)

sns.scatterplot(
    data=index_scores,
    x='Perception_index_mental_health',
    y='experience_index_mental_health',
    hue='RegionCategory',
    palette=color_palette
)

# Add a regression line
sns.regplot(
    data=index_scores,
    x='Perception_index_mental_health',
    y='experience_index_mental_health',
    scatter=False,  # Set to False to avoid plotting data points again
    color='black',  # Set the color of the regression line
    line_kws={"linewidth": 1}  # Set the line width of the regression line
)

countries_to_label = ['United Kingdom','Malta','Jordan','Lebanon','Romania','Paraguay','Burkina Faso','Vietnam','Cameroon','Uruguay','Algeria','Estonia','Belgium','Philippines','China','Peru','Cyprus','Albania','Moldova','United Arab Emirates','Guinea','Japan','Zimbabwe','Venezuela','Namibia','Thailand','France','New Zealand','United States','China','Israel','Afghanistan','South Africa','Bangladesh','Zambia','Canada','Peru','China','Nepal','India','Costa Rica','Mongolia','Hungary','Ghana','France','Finland','Sweden','Denmark','Norway','Israel','India','Afganistan','Mauritius','South Korea','Cyprus','Chile','Brazil','Tanzania','Slovakia','Uzbekistan','Venezuela','Colombia']
#countries_to_label = merged_crime.Country.unique()
# Add country names as labels to the points
for i, row in index_scores.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['Perception_index_mental_health'], row['experience_index_mental_health'], row['Country'],  fontsize=12)

# Set x and y axis labels
plt.xlabel('Mental Health risks Perception index',fontsize=16)
plt.ylabel('Mental Health risks Experience index',fontsize=16)

# Move legend outside plot area
plt.legend(loc='upper left', fontsize=14)

# Display the plot

plt.savefig('./Figures/mental_health_experience_perception.pdf',bbox_inches='tight')
plt.show();


# Create a scatter plot
plt.subplots(figsize=(8, 8), dpi=200)

# Apply the function to create the 'RegionCategory' column
#index_scores['RegionCategory'] = index_scores['GlobalRegion'].apply(map_region_category)

sns.scatterplot(
    data=index_scores,
    x='Perception_index_weather',
    y='experience_index_weather',
    hue='RegionCategory',
    palette=color_palette
)

# Add a regression line
sns.regplot(
    data=index_scores,
    x='Perception_index_weather',
    y='experience_index_weather',
    scatter=False,  # Set to False to avoid plotting data points again
    color='black',  # Set the color of the regression line
    line_kws={"linewidth": 1}  # Set the line width of the regression line
)

countries_to_label = ['United States','United Arab Emirates','Albania','Panama','Latvia','Thailand','Kenya','Cameroon','Ecuado','Denmark','Taiwan','Chile','Romania','Sweden','Pakistan','Malta','Georgia','Jordan','Netherlands','Cyprus','China','Egypt','United Kingdom','Belgium','Australia','Canada','South Korea','Mauritius','Uruguay','Russia','Myanmar','Malaysia','Poland','Kyrgyzstan','Mongolia','Italy','Panama','Sovenia','Germany','Philippines','Afganistan','Zambia','Mexico','India','Sierra Leone','Mali','Burkina Faso','Zambia','Mozambique','Gabon','Guinea','Congo Brazzaville','Nepal','Switzerland','Brazil','Moldova','Indonesia','Costa Rica','Ghana','Benin']
#countries_to_label = index_scores.Country.unique()
# Add country names as labels to the points
for i, row in index_scores.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['Perception_index_weather'], row['experience_index_weather'], row['Country'],  fontsize=12)

# Set x and y axis labels
plt.xlabel('Serious weather risks Perception index',fontsize=16)
plt.ylabel('Serious weather risks Experience index',fontsize=16)

# Move legend outside plot area
plt.legend(loc='upper left', fontsize=14)

# Display the plot

plt.savefig('./Figures/weather_experience_perception.pdf',bbox_inches='tight')
plt.show();


index_scores.head()


merged_first.head()


risk_index_merged = merged_first.merge(index_scores, on='Country', how='inner')
risk_index_merged.head()


import matplotlib as mpl


cmap = mpl.cm.YlOrBr
# Set the size of the plot
fig, ax = plt.subplots(figsize=(8, 8), dpi=150)

# Plot the base map in grey color
regions.plot(ax=ax, facecolor='lightgrey', edgecolor='white', linewidth=0.5)

# Remove grids
ax.grid(False)

# Risk frequency
risk_plot = risk_index_merged.plot(ax=ax, column='risk_impact_index_crime', cmap=cmap, legend=False)

# Create a colorbar within the main graph and make it horizontal
norm = plt.Normalize(risk_index_merged['risk_impact_index_crime'].min(), risk_index_merged['risk_impact_index_crime'].max())
cbar_ax = fig.add_axes([0.25, 0.35, 0.5, 0.02])  # Adjust the position and size of the colorbar
cbar = mcb.ColorbarBase(cbar_ax, cmap=cmap, norm=norm, orientation='horizontal')
cbar.ax.set_xlabel('Average crime risks (%)',fontsize = 6)
cbar.ax.tick_params(labelsize=6)

# Add markers and store annotations in a list
texts = []
for idx, row in risk_index_merged.iterrows():
    ax.scatter(row.geometry.centroid.x, row.geometry.centroid.y, s=3, color='red')
    if (row['globalregion_category'] != 'Eastern Europe') &  (row['globalregion_category'] != 'Southern Europe') :
        texts.append(ax.annotate(row['Country'], xy=(row.geometry.centroid.x, row.geometry.centroid.y), xytext=(5, 5), textcoords='offset points', fontsize=4, color='black'))

# Optimize annotation positions to minimize overlaps
adjust_text(texts, ax=ax)

# Hide the x-axis and y-axis ticks and labels
ax.set_xticks([])
ax.set_yticks([])
#plt.tight_layout()
plt.savefig('./Figures/risk imoact crime_map.pdf')
# Show the plot
plt.show()


cmap = mpl.cm.YlOrBr
# Set the size of the plot
fig, ax = plt.subplots(figsize=(8, 8), dpi=150)

# Plot the base map in grey color
regions.plot(ax=ax, facecolor='lightgrey', edgecolor='white', linewidth=0.5)

# Remove grids
ax.grid(False)

# Risk frequency
risk_plot = risk_index_merged.plot(ax=ax, column='risk_impact_index_crime', cmap=cmap, legend=False)

# Create a colorbar within the main graph and make it horizontal
norm = plt.Normalize(risk_index_merged['risk_impact_index_traffic'].min(), risk_index_merged['risk_impact_index_traffic'].max())
cbar_ax = fig.add_axes([0.25, 0.35, 0.5, 0.02])  # Adjust the position and size of the colorbar
cbar = mcb.ColorbarBase(cbar_ax, cmap=cmap, norm=norm, orientation='horizontal')
cbar.ax.set_xlabel('Average crime risks (%)',fontsize = 6)
cbar.ax.tick_params(labelsize=6)

# Add markers and store annotations in a list
texts = []
for idx, row in risk_index_merged.iterrows():
    ax.scatter(row.geometry.centroid.x, row.geometry.centroid.y, s=3, color='red')
    if (row['globalregion_category'] != 'Eastern Europe') &  (row['globalregion_category'] != 'Southern Europe') :
        texts.append(ax.annotate(row['Country'], xy=(row.geometry.centroid.x, row.geometry.centroid.y), xytext=(5, 5), textcoords='offset points', fontsize=4, color='black'))

# Optimize annotation positions to minimize overlaps
adjust_text(texts, ax=ax)

# Hide the x-axis and y-axis ticks and labels
ax.set_xticks([])
ax.set_yticks([])
#plt.tight_layout()
plt.savefig('./Figures/risk impact traffic_map.pdf')
# Show the plot
plt.show()


df_pop = pd.read_csv("./shapefiles/population.csv")
df_pop.head()


df_pop_2021 = df_pop[df_pop['Year'] == 2021]
df_pop_2021 = df_pop_2021.rename(columns={'Region, subregion, country or area *': 'Country'}) 
df_pop_2021 = df_pop_2021.drop(['Year','Total Population (thousands)'], axis=1)
df_pop_2021.head()


#converting into numeric values
df_pop_2021['Population Density (persons per square km)'] = pd.to_numeric(df_pop_2021['Population Density (persons per square km)'], errors='coerce')
df_pop_2021.head()


df_pop_2021['Country'] = df_pop_2021['Country'].replace('Bolivia (Plurinational State of)', 'Bolivia')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Bosnia and Herzegovina', 'Bosnia Herzegovina')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Congo', 'Congo Brazzaville')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Czechia', 'Czech Republic')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Iran (Islamic Republic of)', 'Iran')
df_pop_2021['Country'] = df_pop_2021['Country'].replace("Côte d'Ivoire", 'Ivory Coast')
df_pop_2021['Country'] = df_pop_2021['Country'].replace("Lao People's Democratic Republic", 'Laos')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Republic of Moldova', 'Moldova')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Russian Federation', 'Russia')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Republic of Korea', 'South Korea')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('United Republic of Tanzania', 'Tanzania')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Türkiye', 'Turkey')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('United Kingdom of Great Britain and Northern Ireland', 'United Kingdom')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('United States of America', 'United States')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Venezuela (Bolivarian Republic of)', 'Venezuela')
df_pop_2021['Country'] = df_pop_2021['Country'].replace('Viet Nam', 'Vietnam')


merged_pop = df_pop_2021.merge(index_scores, on='Country', how='inner')
merged_pop.head()


# Create a scatter plot
sns.scatterplot(data=merged_pop, x='perception_index', y='Population Density (persons per square km)', hue='GlobalRegion')

# Add country names as labels to the points
for i, row in merged_pop.iterrows():
    plt.text(row['perception_index'], row['Population Density (persons per square km)'], row['Country'],  fontsize=8)

# Set x and y axis labels
plt.xlabel('Overall risk perception index')
plt.ylabel('Population Density (persons per square km)')
# set y-axis to log scale
plt.yscale('log') 

# Move legend outside plot area
plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left')

# Display the plot
plt.show();


# Create a scatter plot
sns.scatterplot(data=merged_pop, x='risk_impact_index', y='Population Density (persons per square km)', hue='GlobalRegion')

# Add country names as labels to the points
for i, row in merged_pop.iterrows():
    plt.text(row['risk_impact_index'], row['Population Density (persons per square km)'], row['Country'],  fontsize=8)

# Set x and y axis labels
plt.xlabel('Overall risk impact index')
plt.ylabel('Population Density (persons per square km)')
# set y-axis to log scale
plt.yscale('log') 

# Move legend outside plot area
plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left')

# Display the plot
plt.show();


# Create a scatter plot
sns.scatterplot(data=merged_pop, x='resilience_index', y='Population Density (persons per square km)', hue='GlobalRegion')

# Add country names as labels to the points
for i, row in merged_pop.iterrows():
    plt.text(row['resilience_index'], row['Population Density (persons per square km)'], row['Country'],  fontsize=8)

# Set x and y axis labels
plt.xlabel('Overall resilience index')
plt.ylabel('Population Density (persons per square km)')
# set y-axis to log scale
plt.yscale('log') 
plt.xscale('log') 
# Move legend outside plot area
plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left')

# Display the plot
plt.show();


#Graph for the readyness
# Create a scatter plot
sns.scatterplot(data=merged_pop, x='readiness_index', y='Population Density (persons per square km)', hue='GlobalRegion')

# Add country names as labels to the points
for i, row in merged_pop.iterrows():
    plt.text(row['readiness_index'], row['Population Density (persons per square km)'], row['Country'],  fontsize=8)

# Set x and y axis labels
plt.xlabel('Overall readiness index')
plt.ylabel('Population Density')
# set y-axis to log scale
plt.yscale('log') 

# Move legend outside plot area
plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left')

# Display the plot
plt.show();


index_resi=df.loc[:, ['GlobalRegion','Country', 'Q20', 'Q21','Q10Q11Recode','Q16C', 'Q13','Q16D','Q16A','Q16B']]
index_resi.head()


cols_to_replace = ['Q20', 'Q21','Q10Q11Recode','Q16C', 'Q13','Q16D','Q16A','Q16B']
replace_dict = {98: np.nan, 99: np.nan}
index_resi[cols_to_replace] = index_resi[cols_to_replace].replace(replace_dict)
index_resi.head()


# replace the values 2 with 0.5 and 3 with 0
index_resi.loc[:, ['Q13']] = index_resi.loc[:, ['Q13']].replace({2: 0.5, 3: 0})
index_resi.loc[:, ['Q10Q11Recode']] = index_resi.loc[:, ['Q10Q11Recode']].replace({1: 0.1, 2: 0.2, 3: 0.3, 4: 0.4, 5: 0.5, 6: 0.8, 7: 0.9, 8: 1, 9: 0.7})
index_resi.loc[:, ['Q20','Q16D','Q16A','Q16B','Q16C']] = index_resi.loc[:, ['Q20','Q16D','Q16A','Q16B','Q16C']].replace({3: 0.5, 2: 0})
index_resi.loc[:, ['Q21']] = index_resi.loc[:, ['Q21']].replace({2: 0})
index_resi.head()


# define a function to compute the average of the non-ignored values in each row for perception score (worry) and experience score)
def compute_average(row):
    valid_values = [value for value in row if not pd.isna(value)]
    if len(valid_values) > 0:
        return np.mean(valid_values)
    else:
        return np.nan

# create a new column with the computed average
# index['individual_score'] = index.loc[:, []].apply(compute_average, axis=1)
index_resi['household_score'] = index_resi.loc[:, ['Q21','Q10Q11Recode','Q16C']].apply(compute_average, axis=1)
index_resi['community_score'] = index_resi.loc[:, ['Q13', 'Q16D']].apply(compute_average, axis=1)
index_resi['society_score'] = index_resi.loc[:, ['Q16A','Q16B']].apply(compute_average, axis=1)
index_resi.head()


# Calculating individual resilience score
index_resi['resilience_score'] = index_resi.loc[:, ['Q20','household_score','community_score','society_score']].apply(compute_average, axis=1)
index_resi.head()


resilience_index = index_resi.groupby('Country').agg({'resilience_score': 'mean'}).reset_index()
resilience_index = resilience_index.rename(columns={'resilience_score': 'resilience_index'})
resilience_index.head()


index_scores.head()


index_scores=index_scores.drop('resilience_index',axis=1)


merged_index = resilience_index.merge(index_scores, on='Country', how='inner')
merged_index.head()


# Create a scatter plot
plt.subplots(figsize=(8, 8), dpi=200)

# Apply the function to create the 'RegionCategory' column
merged_index['RegionCategory'] = merged_index['GlobalRegion'].apply(map_region_category)

sns.scatterplot(
    data=merged_index,
    x='resilience_index',
    y='risk_impact_index',
    hue='RegionCategory',
    palette=color_palette
)

# Add a regression line
sns.regplot(
    data=merged_index,
    x='resilience_index',
    y='risk_impact_index',
    scatter=False,  # Set to False to avoid plotting data points again
    color='black',  # Set the color of the regression line
    line_kws={"linewidth": 1}  # Set the line width of the regression line
)

countries_to_label = ['United States','United Arab Emirates','Albania','Panama','Latvia','Thailand','Kenya','Cameroon','Ecuado','Denmark','Taiwan','Chile','Romania','Sweden','Pakistan','Malta','Georgia','Jordan','Netherlands','Cyprus','China','Egypt','United Kingdom','Belgium','Australia','Canada','South Korea','Mauritius','Uruguay','Russia','Myanmar','Malaysia','Poland','Kyrgyzstan','Mongolia','Italy','Panama','Sovenia','Germany','Philippines','Afganistan','Zambia','Mexico','India','Sierra Leone','Mali','Burkina Faso','Zambia','Mozambique','Gabon','Guinea','Congo Brazzaville','Nepal','Switzerland','Brazil','Moldova','Indonesia','Costa Rica','Ghana','Benin','Afghanistan','Albania','Lebanon','Paraguay','Bolivia','Israel','Kyrgyzstan','Panama','Togo','Venezuela','Vietnam','Laos','Cambodia','Malaysia','Uzbekistan']
#countries_to_label = index_scores.Country.unique()
# Add country names as labels to the points
for i, row in merged_index.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['resilience_index'], row['risk_impact_index'], row['Country'],  fontsize=12)

# Set x and y axis labels
plt.xlabel('Resilience Index',fontsize=16)
plt.ylabel('Risk Impact Index',fontsize=16)

# Move legend outside plot area
plt.legend(loc='upper left', fontsize=14)

# Display the plot

plt.savefig('new resilience index vs risk impact.pdf',bbox_inches='tight')
plt.show();


merged_index[merged_index['resilience_index']>0.65]


disaster_ex=df.loc[:, ['GlobalRegion','Country', 'Q18_1','Q17']]
disaster_ex.head()


disaster_ex = disaster_ex[(disaster_ex['Q18_1'] != 98) & (disaster_ex['Q18_1'] != 99)]


disaster_map = {
    1: 'Flood/Heavy rain',
    2: 'Hurricane/Cyclone',
    3: 'Tornado',
    4: 'Thunder or lightning storm',
    5: 'Tsunami',
    6: 'Mudslide',
    7: 'Earthquake',
    8: 'Wildfire',
    9: 'Volcano eruption',
    10: 'Blizzard or snowstorm',
    11: 'Other',
    50: 'Drought',
}

# Use the replace method to replace the numbers with the region names
disaster_ex['Q18_1'] = disaster_ex['Q18_1'].replace(disaster_map)
disaster_ex.head()


# Group the data by 'Q18_1' and count the occurrences
q18_1_counts = disaster_ex['Q18_1'].value_counts()

# Sort the counts in descending order
q18_1_counts = q18_1_counts.sort_values(ascending=True)

# Calculate the percentage for each type of disaster
total = q18_1_counts.sum()
q18_1_percentages = (q18_1_counts / total) * 100

# Sort the percentages in descending order
q18_1_percentages = q18_1_percentages.sort_values(ascending=True)

# Create a horizontal bar plot
plt.barh(q18_1_percentages.index, q18_1_percentages.values)

# Set the labels and title
plt.xlabel('Percentage')
plt.ylabel('Types of disaster')
plt.title('Percentage who experienced each type of disaster')


plt.savefig('./Figures/Disaster_experience.pdf')

# Display the plot
plt.show()


df_disaster = pd.read_csv("./shapefiles/disasters.csv")
df_disaster = df_disaster.rename(columns={'Country name': 'Country'}) 
df_disaster.head()


df_disaster= df_disaster[df_disaster['Year'].isin([2020])]
df_disaster.head()


disaster_ex.head()


disaster_ex_per = disaster_ex.dropna(subset=['Q17'])
disaster_ex_per.head()


# Filter the DataFrame for 'Q17' equal to 1
selected_1 = disaster_ex_per[disaster_ex_per['Q17'] == 1]

# Count the occurrences of each country
country_counts = selected_1['Country'].value_counts()

country_counts_all = disaster_ex_per['Country'].value_counts()
# Calculate the percentage for each country
country_percentage = (country_counts /country_counts_all) * 100

# Create a new DataFrame with 'Country' and 'Percentage' columns
df_percentage_ex = pd.DataFrame({'Country': country_percentage.index, 'Percentage_disaster_ex': country_percentage.values})

# Sort the DataFrame by 'Country' column
df_percentage_ex = df_percentage_ex.sort_values(by='Country')

# Reset the index
df_percentage_ex = df_percentage_ex.reset_index(drop=True)

df_percentage_ex.head()


merged_disaster = df_disaster.merge(df_percentage_ex, on='Country', how='inner')
merged_disaster.head()


merged_disaster = merged_disaster.merge(index_region, on='Country', how='inner')
merged_disaster.head()


# List of countries to label
countries_to_label = ['United Kingdom','Afghanistan','Bangladesh','Dominican Republic','Honduras','Iraq','Kenya','Mali','Namibia','Pakistan','Philippines','South Africa','New Zealand','Australia','United States','Ukraine','Burkina Faso','Costa Rica','Zimbabwe','Namibia','Mexico','Hong Kong','Iran','Cambodia','Armenia','Sierra Leone','United Arab Emirates','Mozambique']

plt.figure(figsize=(8, 8),dpi=150)

# Create the linear regression plot
sns.scatterplot(
    data=merged_disaster,
    x='Number of people affected by disasters per 100,000', 
    y='Percentage_disaster_ex', 
)

# Add country names as labels to the points
for i, row in merged_disaster.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['Number of people affected by disasters per 100,000'], row['Percentage_disaster_ex'], row['Country'], fontsize=10)

# Set x and y axis labels
plt.xlabel('Number of people affected by disasters per 100,000',fontsize=16)
plt.ylabel('% Have experienced a disaster in past five years',fontsize=16)
#plt.xscale('log')
plt.savefig('./Figures/Disaster_experience_vs_actual.pdf')
# Display the plot
plt.show()


merged_res = merged_disaster.merge(resilience_index, on='Country', how='inner')
merged_res.head()


# List of countries to label
countries_to_label = ['United Kingdom','Afghanistan','Bangladesh','Dominican Republic','Honduras','Philippines','Kenya','Mali','Namibia','Pakistan','Philippines','South Africa','New Zealand','Australia','United States','Ukraine','Costa Rica','Zimbabwe','Namibia','Indonesia','Hong Kong','Iran','Cambodia','Armenia','Sierra Leone','United Arab Emirates','Mozambique','Switzerland','Vietnam','Belgium','Norway','Israel','Uzbekistan','Lebanon','Albania','Paraguay','Peru','Benin','Uganda','Uzbekistan','Gabon','Bolivia','Nigeria','Poland','Slovakia','Australia','Slovenia','New Zealand']
#countries_to_label =  merged_res.Country.unique()
plt.figure(figsize=(8, 8),dpi=150)

# Create the linear regression plot
sns.scatterplot(
    data=merged_res,
    x='resilience_index', 
    y='Percentage_disaster_ex', 
)

# Add country names as labels to the points
for i, row in merged_res.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['resilience_index'], row['Percentage_disaster_ex'], row['Country'], fontsize=14)

# Set x and y axis labels
plt.xlabel('Resilience index', fontsize=16)
plt.ylabel('% Have experienced a disaster in past five years',fontsize=16)

plt.savefig('./Figures/Disaster_experience_vs_resilience.pdf')
# Display the plot
plt.show()


filtered_data = merged_res[(merged_res['resilience_index'] < 0.4) & (merged_res['Death rates from disasters'] <0.05)]
filtered_data


# List of countries to label
countries_to_label = ['United Kingdom','Afghanistan','Belgium','Bolivia','Canada','El Salvador','France','Honduras','Mongolia','Nepal','Netherlands','Pakistan','New Zealand','Australia','United States','Panama','Philippines','South Africa','Uganda','Venezuela','Afghanistan','Vietnam','Switzerland','Laos','Bolivia','Paraguay','Lebanon','']

plt.figure(figsize=(10, 10),dpi=150)

# Create the linear regression plot
sns.regplot(
    data=merged_res,
    x='resilience_index', 
    y='Death rates from disasters', 
)

# Add country names as labels to the points
for i, row in merged_res.iterrows():
    if row['Country'] in countries_to_label:
        plt.text(row['resilience_index'], row['Death rates from disasters'], row['Country'], fontsize=10)

# Set x and y axis labels
plt.xlabel('Resilience index')
plt.ylabel('Death rate from disasters')

plt.savefig('./Figures/Disaster_actual_vs_resilience.pdf')
# Display the plot
plt.show()

	COUNTRY	geometry
0	Afghanistan	POLYGON ((61.27655 35.60725, 61.29638 35.62854...
1	Albania	POLYGON ((19.57083 41.68527, 19.58195 41.69569...
2	Algeria	POLYGON ((4.60335 36.88791, 4.63555 36.88638, ...
3	American Samoa	POLYGON ((-170.74390 -14.37555, -170.74942 -14...
4	Andorra	POLYGON ((1.44584 42.60194, 1.48653 42.65042, ...
...	...	...
246	Vietnam	MULTIPOLYGON (((107.07896 17.10804, 107.08333 ...
247	Wallis and Futuna	MULTIPOLYGON (((-178.06082 -14.32389, -178.137...
248	Yemen	MULTIPOLYGON (((47.25445 13.61528, 47.16888 13...
249	Zambia	POLYGON ((30.21302 -14.98172, 30.21916 -15.096...
250	Zimbabwe	POLYGON ((32.48887 -21.34445, 32.46541 -21.325...

	Country	Q2_1	GlobalRegion
1	United States	7.0	6.0
2	United States	3.0	6.0
3	United States	3.0	6.0
4	United States	10.0	6.0
5	United States	1.0	6.0
...	...	...	...
125905	Serbia	7.0	12.0
125907	Serbia	10.0	12.0
125908	Serbia	5.0	12.0
125909	Serbia	7.0	12.0
125910	Serbia	18.0	12.0

	Country	Q2_1	GlobalRegion	risk_category
1	United States	7.0	6.0	HEALTH3
2	United States	3.0	6.0	CRIME
3	United States	3.0	6.0	CRIME
4	United States	10.0	6.0	ECONOMY
5	United States	1.0	6.0	TRANS1
...	...	...	...	...
125905	Serbia	7.0	12.0	HEALTH3
125907	Serbia	10.0	12.0	ECONOMY
125908	Serbia	5.0	12.0	HEALTH1
125909	Serbia	7.0	12.0	HEALTH3
125910	Serbia	18.0	12.0	ENVIRONMENT1

	GlobalRegion	Country	Q4A	Q4B	Q4C	Q4D	Q4E	Q4F	Q4G	Q5A	Q5B	Q5C	Q5D	Q5E	Q5F	Q5G	Q20
0	6.0	United States	1.0	1.0	1.0	1.0	2.0	3.0	NaN	4.0	4.0	4.0	4.0	4.0	4.0	4.0	1.0
1	6.0	United States	2.0	2.0	2.0	3.0	3.0	3.0	3.0	4.0	4.0	4.0	4.0	4.0	4.0	4.0	1.0
2	6.0	United States	3.0	3.0	1.0	2.0	1.0	3.0	NaN	4.0	4.0	4.0	4.0	4.0	4.0	4.0	1.0
3	6.0	United States	3.0	3.0	3.0	2.0	3.0	3.0	NaN	4.0	4.0	4.0	4.0	2.0	2.0	4.0	1.0
4	6.0	United States	3.0	3.0	2.0	3.0	2.0	3.0	3.0	2.0	4.0	4.0	4.0	2.0	2.0	4.0	1.0

	GlobalRegion	Country	Q4A	Q4B	Q4C	Q4D	Q4E	Q4F	Q4G	Q5A	Q5B	Q5C	Q5D	Q5E	Q5F	Q5G	Q20
0	6.0	United States	1.0	1.0	1.0	1.0	2.0	3.0	NaN	4.0	4.0	4.0	4.0	4.0	4.0	4.0	1.0
1	6.0	United States	2.0	2.0	2.0	3.0	3.0	3.0	3.0	4.0	4.0	4.0	4.0	4.0	4.0	4.0	1.0
2	6.0	United States	3.0	3.0	1.0	2.0	1.0	3.0	NaN	4.0	4.0	4.0	4.0	4.0	4.0	4.0	1.0
3	6.0	United States	3.0	3.0	3.0	2.0	3.0	3.0	NaN	4.0	4.0	4.0	4.0	2.0	2.0	4.0	1.0
4	6.0	United States	3.0	3.0	2.0	3.0	2.0	3.0	3.0	2.0	4.0	4.0	4.0	2.0	2.0	4.0	1.0

Step 1: Simple Exploratory Data Analysis¶

Step 1.1. Just some simple bar charts¶

Step 1.2: Spider chart of each region perception to risks and experience to disasters¶

Step 1.3: Lets compare the gap between what people thinks with what they actually experienced¶

Step 2: Plot country statistics on a map¶

Step 2. Plot the 3 main risks on a global map¶

Step 2.1. Transport risks, experience and resilience¶

2.1.1 Transport perception map¶

2.1.2. Visualise the ranked map of transport experience¶

Step 2.1.3: Visualise the resilience of people with transport risks¶

Step 2.2: Visualise a heat map of crime risks¶

2.2.2. Visualise the ranked map of crime experience¶

Step 3.1 Selecting most worried risks by people¶

Step 3.2 Calculating Perception Index, Risk Impact Index, and Resilience Index¶

For given risk (eg: crime)¶

Resilience index for all seven risks¶

Step 3.3. Comparison of mean impact index from each region.¶

Step 3.4. Comparison with traffic statistical data¶

Step 3.5 Traffic perception vs Traffic Experience index¶

Step 3.6. Traffic impact vs Actual¶

Step 3.7 Comparison with crime statistical data¶

Step 3.8 Crime perception vs Crime Experience¶

Step 3.9 Crime impact vs actual¶

Step 3.10: Mental health and Serious Weather perception vs experience index¶

Make a heat map of Risk Impact index for Traffic and Crime¶

Step 4: Comparison of overall risk perception, impact and resilient index with population statistical data¶

New Resilience index¶

Comparison between New Resilience Index vs Risk Impact Index¶

Peolple experience on natural disasters and country's resilience to disasters¶

Experience with Disasters¶

Comparison with disaster statistical data¶

people experience disaster in past 5 yrs vs actual data¶

	WPID_RANDOM	WP5	Country	WGT	Year	GlobalRegion	country.in.both.waves	CountryIncomeLevel2021	CountryIncomeLevel2019	PROJWT_2021	...	REGION_ZAF	REGION2_ZMB	REGION_ZWE	Q2_1_2019A	Q2_1_2019B	Q5A_2019	Q5B_2019	Q5C_2019	Q5D_2019	Q5F_2019
0	174262180.0	1.0	United States	2.132627	2021.0	6.0	1.0	4.0	NaN	562058.627707	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
1	115327451.0	1.0	United States	0.620385	2021.0	6.0	1.0	4.0	NaN	163503.920074	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2	189972125.0	1.0	United States	0.610080	2021.0	6.0	1.0	4.0	NaN	160787.958907	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
3	173358289.0	1.0	United States	0.490774	2021.0	6.0	1.0	4.0	NaN	129344.456130	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4	183251831.0	1.0	United States	1.910021	2021.0	6.0	1.0	4.0	NaN	503390.265930	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

	CONTINENT	geometry
0	Asia	MULTIPOLYGON (((93.27554 80.26361, 93.31304 80...
1	North America	MULTIPOLYGON (((-25.28167 71.39166, -25.32889 ...
2	Europe	MULTIPOLYGON (((58.06138 81.68776, 57.98055 81...
3	Africa	MULTIPOLYGON (((0.69465 5.77337, 0.66667 5.803...
4	South America	MULTIPOLYGON (((-81.71306 12.49028, -81.72014 ...
5	Oceania	MULTIPOLYGON (((-177.39334 28.18416, -177.3958...
6	Australia	MULTIPOLYGON (((142.27997 -10.26556, 142.21053...

	Country	Q2_1	GlobalRegion	risk_category	globalregion_category
1	United States	7.0	6.0	HEALTH3	Northern America
2	United States	3.0	6.0	CRIME	Northern America
3	United States	3.0	6.0	CRIME	Northern America
4	United States	10.0	6.0	ECONOMY	Northern America
5	United States	1.0	6.0	TRANS1	Northern America
...	...	...	...	...	...
125905	Serbia	7.0	12.0	HEALTH3	Eastern Europe
125907	Serbia	10.0	12.0	ECONOMY	Eastern Europe
125908	Serbia	5.0	12.0	HEALTH1	Eastern Europe
125909	Serbia	7.0	12.0	HEALTH3	Eastern Europe
125910	Serbia	18.0	12.0	ENVIRONMENT1	Eastern Europe

	Country	geometry	globalregion_category	Q5E
0	Afghanistan	POLYGON ((61.27655 35.60725, 61.29638 35.62854...	South Asia	0.327856
1	Albania	POLYGON ((19.57083 41.68527, 19.58195 41.69569...	Southern Europe	0.102307
2	Algeria	POLYGON ((4.60335 36.88791, 4.63555 36.88638, ...	North Africa	0.311801
3	Argentina	MULTIPOLYGON (((-71.85916 -41.01128, -71.83806...	Latin America & Caribbean	0.210000
4	Armenia	MULTIPOLYGON (((46.54037 38.87559, 46.51639 38...	Central Asia	0.166835

	Mean	SD
Risks
Traffic	1.851053	0.782952
Crime	1.928503	0.829155
Food	2.182905	0.777354
Water	2.303742	0.790337
Environment 2	1.913353	0.804176
Health 4	2.256655	0.806659
Work	2.361072	0.757320

	GlobalRegion	Country	Q4A	Q4B	Q4C	Q4D	Q4E	Q4G	Q5A	Q5E	Q5F	Q20	overall_perception_score	overall_experience_score
0	6.0	United States	1.0	1.0	1.0	1.0	0.5	NaN	0.0	0.0	0.0	1.0	0.750000	0.000000
1	6.0	United States	0.5	0.5	0.5	0.0	0.0	0.0	0.0	0.0	0.0	1.0	0.214286	0.000000
2	6.0	United States	0.0	0.0	1.0	0.5	1.0	NaN	0.0	0.0	0.0	1.0	0.416667	0.000000
3	6.0	United States	0.0	0.0	0.0	0.5	0.0	NaN	0.0	0.5	0.5	1.0	0.083333	0.142857
4	6.0	United States	0.0	0.0	0.5	0.0	0.5	0.0	0.5	0.5	0.5	1.0	0.142857	0.214286

	Country	perception_index	experience_index	risk_impact_index	readiness_index	Perception_index_food	Perception_index_water	Perception_index_crime	Perception_index_weather	Perception_index_traffic	...	experience_index_traffic	experience_index_mental_health	experience_index_work	risk_impact_index_food	risk_impact_index_water	risk_impact_index_crime	risk_impact_index_weather	risk_impact_index_traffic	risk_impact_index_mental_health	risk_impact_index_work
0	Afghanistan	0.504098	0.350008	0.217335	0.341858	0.385155	0.345190	0.736069	0.549231	0.474846	...	0.327856	0.439524	0.323498	0.186173	0.175805	0.356339	0.255847	0.207132	0.342541	0.152985
1	Albania	0.466912	0.066757	0.041336	0.239796	0.556056	0.494494	0.385743	0.415075	0.650505	...	0.102307	0.065010	0.042169	0.059362	0.039653	0.025583	0.062248	0.079833	0.047734	0.027612
2	Algeria	0.417045	0.179415	0.086729	0.375000	0.310521	0.324695	0.637205	0.477273	0.517116	...	0.311417	0.093910	0.123457	0.126156	0.094584	0.136694	0.074395	0.200052	0.049208	0.084452
3	Argentina	0.461948	0.152959	0.085949	0.567368	0.413741	0.410410	0.610611	0.533033	0.597793	...	0.210000	0.162000	0.141109	0.053860	0.081456	0.108952	0.161949	0.145457	0.087212	0.067838
4	Armenia	0.322071	0.090433	0.039543	0.574682	0.456609	0.273092	0.236298	0.357287	0.405961	...	0.166835	0.085347	0.061235	0.091223	0.047980	0.015479	0.040265	0.101498	0.052810	0.033333

	Country	GlobalRegion
0	United States	Northern America
1	Egypt	North Africa
2	Morocco	North Africa
3	Lebanon	Middle East
4	Saudi Arabia	Middle East

	Country	Year	Estimated number of road traffic deaths	Estimated road traffic death rate (per 100 000 population)
0	Afghanistan	2019	6033 [5227-6846]	15.9 [13.7-18.0]
1	Afghanistan	2018	5295 [4578-6017]	14.2 [12.3-16.2]
2	Afghanistan	2017	5490 [4737-6246]	15.1 [13.1-17.2]
3	Afghanistan	2016	5240 [4512-5970]	14.8 [12.8-16.9]
4	Afghanistan	2015	4942 [4248-5637]	14.4 [12.3-16.4]

	Country	Year	Estimated number of road traffic deaths	Estimated road traffic death rate (per 100 000 population)	perception_index	experience_index	risk_impact_index	readiness_index	Perception_index_food	Perception_index_water	...	experience_index_work	risk_impact_index_food	risk_impact_index_water	risk_impact_index_crime	risk_impact_index_weather	risk_impact_index_traffic	risk_impact_index_mental_health	risk_impact_index_work	GlobalRegion	resilience_index
0	Afghanistan	2019	6033	15.9	0.504098	0.350008	0.217335	0.341858	0.385155	0.345190	...	0.323498	0.186173	0.175805	0.356339	0.255847	0.207132	0.342541	0.152985	South Asia	0.074298
1	Albania	2019	337	11.7	0.466912	0.066757	0.041336	0.239796	0.556056	0.494494	...	0.042169	0.059362	0.039653	0.025583	0.062248	0.079833	0.047734	0.027612	Southern Europe	0.009912
2	Algeria	2019	8996	20.9	0.417045	0.179415	0.086729	0.375000	0.310521	0.324695	...	0.123457	0.126156	0.094584	0.136694	0.074395	0.200052	0.049208	0.084452	North Africa	0.032523
3	Argentina	2019	6298	14.1	0.461948	0.152959	0.085949	0.567368	0.413741	0.410410	...	0.141109	0.053860	0.081456	0.108952	0.161949	0.145457	0.087212	0.067838	Latin America & Caribbean	0.048765
4	Armenia	2019	590	20.0	0.322071	0.090433	0.039543	0.574682	0.456609	0.273092	...	0.061235	0.091223	0.047980	0.015479	0.040265	0.101498	0.052810	0.033333	Central Asia	0.022725

	Country	crimeIndex
0	Venezuela	83.76
1	Papua New Guinea	80.79
2	South Africa	76.86
3	Afghanistan	76.31
4	Honduras	74.54

	Country	crimeIndex	perception_index	experience_index	risk_impact_index	readiness_index	Perception_index_food	Perception_index_water	Perception_index_crime	Perception_index_weather	...	experience_index_work	risk_impact_index_food	risk_impact_index_water	risk_impact_index_crime	risk_impact_index_weather	risk_impact_index_traffic	risk_impact_index_mental_health	risk_impact_index_work	GlobalRegion	resilience_index
0	Venezuela	83.76	0.532048	0.161137	0.101761	0.492394	0.532000	0.582500	0.711924	0.632000	...	0.157250	0.095095	0.114375	0.181112	0.187813	0.103729	0.053428	0.088100	Latin America & Caribbean	0.050106
1	Guinea	80.79	0.720953	0.246974	0.189689	0.492212	0.680303	0.625755	0.723698	0.774872	...	0.231814	0.164412	0.125000	0.182561	0.232458	0.320624	0.221373	0.203608	Central/Western Africa	0.093367
2	South Africa	76.86	0.589946	0.191910	0.131190	0.342375	0.422696	0.457269	0.838077	0.646707	...	0.160820	0.100549	0.107601	0.290675	0.136307	0.217786	0.105020	0.111085	Southern Africa	0.044916
3	Afghanistan	76.31	0.504098	0.350008	0.217335	0.341858	0.385155	0.345190	0.736069	0.549231	...	0.323498	0.186173	0.175805	0.356339	0.255847	0.207132	0.342541	0.152985	South Asia	0.074298
4	Honduras	74.54	0.421881	0.116495	0.067771	0.583247	0.364286	0.278846	0.563572	0.596386	...	0.109909	0.054342	0.042219	0.096193	0.197621	0.095998	0.041925	0.055856	Latin America & Caribbean	0.039527

	Region, subregion, country or area *	Year	Total Population (thousands)	Population Density (persons per square km)
0	WORLD	1950.0	2 499 322	19.2
1	WORLD	1951.0	2 543 130	19.5
2	WORLD	1952.0	2 590 271	19.9
3	WORLD	1953.0	2 640 279	20.2
4	WORLD	1954.0	2 691 979	20.6

	Country	Population Density (persons per square km)	perception_index	experience_index	risk_impact_index	readiness_index	Perception_index_food	Perception_index_water	Perception_index_crime	Perception_index_weather	...	experience_index_work	risk_impact_index_food	risk_impact_index_water	risk_impact_index_crime	risk_impact_index_weather	risk_impact_index_traffic	risk_impact_index_mental_health	risk_impact_index_work	GlobalRegion	resilience_index
0	Kenya	91.3	0.550674	0.225890	0.147029	0.450754	0.486500	0.449651	0.656627	0.564000	...	0.252014	0.135870	0.149448	0.185790	0.135617	0.234944	0.129289	0.166845	Eastern Africa	0.066274
1	Mauritius	639.9	0.598781	0.090805	0.061365	0.804954	0.403749	0.257360	0.800916	0.767241	...	0.127264	0.045256	0.019750	0.042733	0.096713	0.127411	0.066259	0.072767	Eastern Africa	0.049396
2	Mozambique	40.8	0.704409	0.259018	0.198115	0.596070	0.720226	0.593909	0.681967	0.752160	...	0.235403	0.238552	0.176579	0.172565	0.263672	0.227298	0.213913	0.213046	Eastern Africa	0.118090
3	Uganda	229.5	0.524804	0.242063	0.144261	0.439716	0.409228	0.404810	0.613956	0.612563	...	0.245720	0.120599	0.137814	0.174269	0.189451	0.238705	0.112206	0.145592	Eastern Africa	0.063434
4	Tanzania	71.8	0.480851	0.168458	0.097924	0.507708	0.243927	0.381168	0.527692	0.598684	...	0.150963	0.041667	0.105183	0.128106	0.131994	0.192464	0.135093	0.079215	Eastern Africa	0.049717

	Country	Population Density (persons per square km)
71	WORLD	60.7
144	Sub-Saharan Africa	52.0
216	Northern Africa and Western Asia	43.7
288	Central and Southern Asia	199.9
360	Eastern and South-Eastern Asia	144.1

	GlobalRegion	Country	Q20	Q21	Q10Q11Recode	Q16C	Q13	Q16D	Q16A	Q16B	household_score	community_score	society_score
0	6.0	United States	1.0	1.0	0.8	1.0	1.0	0.0	0.0	1.0	0.933333	0.50	0.5
1	6.0	United States	1.0	1.0	0.8	1.0	1.0	1.0	0.0	1.0	0.933333	1.00	0.5
2	6.0	United States	1.0	1.0	1.0	1.0	0.5	1.0	1.0	1.0	1.000000	0.75	1.0
3	6.0	United States	1.0	0.0	0.5	1.0	1.0	1.0	0.0	1.0	0.500000	1.00	0.5
4	6.0	United States	1.0	1.0	1.0	1.0	0.5	1.0	1.0	1.0	1.000000	0.75	1.0

	Country	resilience_index
0	Afghanistan	0.286347
1	Albania	0.300214
2	Algeria	0.376750
3	Argentina	0.403610
4	Armenia	0.445351

	Country	resilience_index	perception_index	experience_index	risk_impact_index	readiness_index	Perception_index_food	Perception_index_water	Perception_index_crime	Perception_index_weather	...	experience_index_work	risk_impact_index_food	risk_impact_index_water	risk_impact_index_crime	risk_impact_index_weather	risk_impact_index_traffic	risk_impact_index_mental_health	risk_impact_index_work	GlobalRegion	RegionCategory
7	Bangladesh	0.669111	0.549897	0.164906	0.108355	0.617363	0.535897	0.350709	0.495699	0.648374	...	0.167903	0.095999	0.045969	0.075647	0.203635	0.171375	0.119512	0.136769	South Asia	Asia
15	Cambodia	0.686195	0.559931	0.180111	0.114825	0.716981	0.641206	0.402414	0.412616	0.677094	...	0.193044	0.132836	0.043099	0.057552	0.204914	0.206571	0.148101	0.122527	South-eastern Asia	Asia
45	Indonesia	0.741710	0.520312	0.132280	0.076742	0.768744	0.322243	0.208136	0.618147	0.730076	...	0.165722	0.052183	0.025977	0.072106	0.149332	0.169719	0.055662	0.094526	South-eastern Asia	Asia
59	Laos	0.730007	0.378373	0.119874	0.056866	0.631823	0.346074	0.245902	0.330189	0.499482	...	0.134458	0.074869	0.040798	0.027569	0.106540	0.120531	0.035056	0.063302	South-eastern Asia	Asia
63	Malaysia	0.681243	0.487752	0.137288	0.079151	0.612193	0.391457	0.374122	0.584677	0.595046	...	0.145960	0.061157	0.057967	0.067682	0.099049	0.119504	0.080724	0.077160	South-eastern Asia	Asia
76	New Zealand	0.657860	0.279356	0.100482	0.037319	0.704661	0.134500	0.145646	0.390500	0.366000	...	0.088516	0.013000	0.012776	0.046500	0.057500	0.072666	0.103496	0.049655	Australia and New Zealand	Oceania
80	Norway	0.666422	0.211513	0.079543	0.022601	0.785132	0.170170	0.080160	0.156657	0.273774	...	0.071750	0.008884	0.002004	0.013778	0.021522	0.061250	0.102056	0.023438	Northern/Western Europe	Europe
85	Philippines	0.753132	0.625205	0.264887	0.180194	0.911824	0.593093	0.369739	0.682182	0.783000	...	0.208750	0.154780	0.101703	0.132007	0.386625	0.267660	0.184434	0.149485	South-eastern Asia	Asia
94	Singapore	0.652978	0.259689	0.072018	0.024985	0.629727	0.251000	0.156344	0.214357	0.241466	...	0.080339	0.024673	0.004267	0.011237	0.013609	0.060746	0.074601	0.036851	South-eastern Asia	Asia
102	Switzerland	0.732229	0.277450	0.141517	0.040082	0.699161	0.276500	0.188188	0.313814	0.449850	...	0.205455	0.022545	0.019539	0.026652	0.105567	0.028682	0.060060	0.044480	Northern/Western Europe	Europe
112	United Arab Emirates	0.690991	0.174034	0.066737	0.021267	0.593326	0.168495	0.110558	0.158026	0.166833	...	0.086653	0.022103	0.011092	0.012538	0.016141	0.096916	0.025625	0.042489	Middle East	Asia
114	United States	0.671902	0.308489	0.121286	0.054086	0.778443	0.230846	0.200995	0.414257	0.400995	...	0.084328	0.031530	0.013433	0.066866	0.121393	0.101843	0.105184	0.050749	Northern America	North America
116	Uzbekistan	0.674516	0.218663	0.043214	0.016341	0.681667	0.125752	0.172189	0.169534	0.308824	...	0.078125	0.010111	0.017281	0.010152	0.022970	0.036548	0.028943	0.025709	Central Asia	Asia
118	Vietnam	0.712650	0.451059	0.170166	0.092604	0.840081	0.548259	0.373380	0.439182	0.483085	...	0.215319	0.107445	0.052448	0.064247	0.129358	0.180389	0.116169	0.093750	South-eastern Asia	Asia

	Country	Year
0	Afghanistan	1900
1	Afghanistan	1910
2	Afghanistan	1920
3	Afghanistan	1930
4	Afghanistan	1940

	Country	Year	Death rates from disasters	Number of people affected by disasters per 100,000
12	Afghanistan	2020	1.700606	9519.594000
25	Africa	2020	0.156916	2562.882600
38	Albania	2020	0.000000	0.980835
51	Algeria	2020	0.110379	50.602490
64	American Samoa	2020	0.000000	0.000000

	Country	Percentage_disaster_ex
0	Afghanistan	45.573441
1	Albania	30.730731
2	Algeria	7.615230
3	Argentina	19.980020
4	Armenia	47.887324

Step 1: Simple Exploratory Data Analysis¶

Step 1.1. Just some simple bar charts¶

Step 1.2: Spider chart of each region perception to risks and experience to disasters¶

Step 1.3: Lets compare the gap between what people thinks with what they actually experienced¶

Step 2: Plot country statistics on a map¶

Step 2. Plot the 3 main risks on a global map¶

Step 2.1. Transport risks, experience and resilience¶

2.1.1 Transport perception map¶

2.1.2. Visualise the ranked map of transport experience¶

Step 2.1.3: Visualise the resilience of people with transport risks¶

Step 2.2: Visualise a heat map of crime risks¶

2.2.2. Visualise the ranked map of crime experience¶

Step 3: Make risk-related indexes¶

Step 3.1 Selecting most worried risks by people¶

Step 3.2 Calculating Perception Index, Risk Impact Index, and Resilience Index¶

For given risk (eg: crime)¶

Resilience index for all seven risks¶

Step 3.3. Comparison of mean impact index from each region.¶

Step 3.4. Comparison with traffic statistical data¶

Step 3.5 Traffic perception vs Traffic Experience index¶

Step 3.6. Traffic impact vs Actual¶

Step 3.7 Comparison with crime statistical data¶

Step 3.8 Crime perception vs Crime Experience¶

Step 3.9 Crime impact vs actual¶

Step 3.10: Mental health and Serious Weather perception vs experience index¶

Make a heat map of Risk Impact index for Traffic and Crime¶

Step 4: Comparison of overall risk perception, impact and resilient index with population statistical data¶

New Resilience index¶

Comparison between New Resilience Index vs Risk Impact Index¶

Peolple experience on natural disasters and country's resilience to disasters¶

Experience with Disasters¶

Comparison with disaster statistical data¶

people experience disaster in past 5 yrs vs actual data¶